From the collection of the
7 n
z m
o Prelinger
i a
v JJibrary
San Francisco, California
2006
UNIVERSITY OF CALIFORNIA PUBLICATIONS
IN
CULTURE AND SOCIETY
VOLUME III
EDITORS
R. L. BEALS
FRANKLIN FEARING
W. S. ROBINSON
TVA AND THE GRASS ROOTS
A Study in the Sociology of Formal
Organization
BY
PHILIP SELZNICK
UNIVERSITY OF CALIFORNIA PRESS
BERKELEY AND LOS ANGELES
1949
UNIVERSITY OF CALIFORNIA PUBLICATIONS IN CULTURE AND SOCIETY
EDITORS : R. L. BEALS, FRANKLIN FEARING, W. S. EOBINSON
Volume 3
Submitted by editors February 23, 1948
Issued April 1, 1949
Price: cloth, $3.75; paper, $2.75
UNIVERSITY OF CALIFORNIA PRESS
BERKELEY AND Los ANGELES
CALIFORNIA
CAMBRIDGE UNIVERSITY PRESS
LONDON, ENGLAND
PRINTED IN THE UNITED STATES OF AMERICA
PREFACE
THIS STUDY was made possible by a field fellowship granted me in 1942-
1943 by the Social Science Research Council ; and by the cooperation
of the Tennessee Valley Authority, which opened its doors to scientific
inquiry "with no strings attached." To both of these organizations I am
indebted for the opportunity they created.
The materials of this inquiry were gathered during 1942-1943. The
analysis was not committed to paper until three years later a wholly
incidental consequence of the war years. Subsequent developments in
the TVA program and organization have not been taken into account.
But since the primary interest of the study is in theoretical considera-
tions, the delay is not, perhaps, as consequential as it might otherwise
be. On the other hand, the situation within TVA as it was in 1943 repre-
sented the close of a decade of its operation, a point to be borne in mind
by those interested in TVA's history for its own sake.
The files and the personnel of the Tennessee Valley Authority were
the primary sources of research data. The unpublished "record" has been
accorded the same status as personal interview materials, so that sources
and quotations cannot always be given specific reference. I have en-
deavored to protect the anonymity of those in and out of the Authority
who have helped me to an understanding of the TVA's methods and pro-
gram. At the same time, informants on questions of detail have been
restricted to those within TVA who have worked on the programs dis-
cussed. A check with the written record was made wherever possible.
Interviews with officials in Washington and in the Tennessee Valley
states were also of assistance.
It is hoped that a contribution has been made here toward the evolu-
tion of a theory of organization. In that sense, the study is not practical
or programmatic. It is believed, however, that a practical relevance will
be discerned by those involved in action who must take into account such
general relations within and among organizations as are studied here.
It must also be emphasized that what is presented here is only one aspect
of the total TVA picture. For more general presentations of the Author-
ity's program, the reader is referred to such volumes as David E. Lilien-
thal's TVA: Democracy on the March, C. Herman Pritchett's The
Tennessee Valley Authority: A Study in Public Administration, and
Herman Finer's TVA : Lessons for International Application.
It is unfortunate that the nature of the materials makes impossible
explicit acknowledgment of my debt to the many individuals who gave
[v]
vi Preface
liberally of time and faith so that I might have the materials for a
realistic analysis. Without the guidance which only participants can
give, much of this analysis could never have been made explicit. In addi-
tion to those unnamed, I wish to thank Daniel Bell, Edmund de S.
Brunner, Patterson H. French, Max M. Kampelman, Robert S. Lynd,
Robert K. Merton, and John D. Millett. To the critical intelligence of
Gertrude Jaeger there is a special obligation.
p. s.
Los Angeles
November, 1947
CONTENTS
INTRODUCTION
TVA and Democratic Planning 3
PART ONE : OFFICIAL DOCTRINE PARAPHRASE
AND ANALYSIS
I. The Idea of a "Grass Boots" Administration 19
The need for decentralization of administration .... 22
Minimum essentials 28
Managerial autonomy 29
The partnership of TVA and the people's institutions . . 37
Decentralization and regional unity 41
II. The Functions and Dilemmas of Official Doctrine .... 47
Sources and functions 47
Unanalyzed abstractions 59
Administrative discretion 64
Inherent dilemmas 69
Implications f or U.S.-TVA relations 74
PART Two : THE INFORMAL COOPTATION OF GRASS-ROOTS
ELEMENTS AN ADMINISTRATIVE CONSTITUENCY
AND ITS CONSEQUENCES
III. TVA and the Farm Leadership : The Construction
of an Administrative Constituency 85
TVA's agricultural responsibilities 85
Approaching the grass roots 91
The memorandum of understanding 95
The decision for phosphates 98
Extent of the program 99
The TVA machinery 104
IV. TVA and the Farm Leadership (Continued) 117
Character of the extension service 117
Pattern of cooperation 124
Integration of TVA and extension-service programs . . 129
Relations to the Farm Bureau 141
An administrative constituency 145
[vii]
viii Contents
V. Unanticipated Consequences, 1 : The Struggle in
Agriculture and the Role of TVA .* , . . . . ... 155
Interagency rivalries in agriculture 157
TVA and the Farm Security Administration 164
The Soil Conservation Service crisis ..... , .:,. .... . . 169
VI. Unanticipated Consequences, 2 : Land-Use Policy
and the Character of TVA . . . ,; : . 181
TVA and public lands 186
Conservation goals in the early period 190
The reservoir protective strip 193
A reversal of policy 202
"Fanaticism" and administrative decision 205
PART THREE : THE FORMAL COOPTATION OF GRASS-ROOTS
ELEMENTS AMBIGUITIES OF PARTICIPATION
VII. The Voluntary Association at the End Point
of Administration 217
Formal cooptation and agricultural democracy .... 219
Evolution of TVA interest in cooperatives 226
Cooptation in the fertilizer program 230
Rural electrification cooperatives 238
Other voluntary associations 242
CONCLUSION
Guiding Principles and Interpretation : A Summary 249
Sociological directives 250
Unanticipated consequences in organized action 253
The cooptative mechanism 259
Empirical argument restated v ..... 262
Implications for democratic planning 264
Published Sources . 267
TABLES
1. Distribution of TVA fertilizers to January 1, 1943 100
2. Summary of payments under cooperative contracts with
state and local institutions 101
3. Eeimbursements made to land-grant colleges under
cooperative agreements (1935-1943) 101
4. Personnel on cooperative TVA-land-grant college agricultural
program, salaries reimbursed by TVA in 1942 102
5. TVA materials investigated by experiment stations
in the Tennessee Valley to 1943 103
6. Test-demonstration farms, as of January 1, 1943 103
7. Number of families removed from TVA reservoir
areas to March 1, 1943 104
8. Allotment of funds, by program and method of disbursement within
TVA Agricultural Eelations Department (1943-1944 proposals) . . . Ill
9. Summary of assisting citizens in U. S. agricultural programs (1939) . . . 223
CHARTS
1. Tennessee Valley Authority organization 106
2. TVA units having agricultural responsibilities 107
3. TVA Department of Agricultural Eelations " . . 110
Introduction
TVA AND DEMOCEATIC PLANNING
INTRODUCTION
TVA AND DEMOCRATIC PLANNING
In this country we are very vain of our political institutions, which are singular in
this, that they sprung, within the memory of living men, from the character and
condition of the people, which they still express with sufficient fidelity . . .
EMERSON
WHATEVER the ultimate outcome, it is evident that modern society has
already moved rather far into the age of control. It is an age marked by
widening efforts to master a refractory industrial system. That a tech-
nique for control will emerge, that there is and will be planning, is
hardly in question. "What is more doubtful is the character and direction
of the new instruments of intervention and constraint. For these have
been born of social crisis, set out piecemeal as circumstances have de-
manded ; they have not come to us as part of a broad and conscious vision.
As a consequence, the foundations of a clear-cut choice between totali-
tarian and democratic planning have not been adequately laid; nor has
the distinction been altogether clear between planning directed toward
some acceptable version of the common good and planning for the effec-
tive maintenance of existing and emerging centers of privilege and
power.
Democracy has to do with means, with instruments, with tools which
define the relation between authority and the individual. In our time,
new and inescapable tasks demand a choice among available means
within the framework of increased governmental control. It is therefore
especially important to examine those organizations which are proposed
as contributions to the technique of democratic planning. An example
of such a proposed contribution is the Tennessee Valley Authority.
On June 25, 1942, The Times (London) published a brief review of
TVA under the heading "The Technique of Democratic Planning." The
Times correspondent reported that he was impressed by the physical
accomplishments of dam and power plant construction, but what inter-
ested him most was "the technique which the TVA had adopted with
the deliberate aim of reconciling over-all planning with the values of
democracy." Here The Times reflected what many feel to be the enduring
significance of this much discussed government agency. The theme of
democracy in government administration was also prominent in a widely
distributed book, TVA: Democracy on the March, written by David E.
Lilienthal, and in numerous speeches and pamphlets emanating from
[3]
4 University of California Publications in Culture and Society
the Authority. In addition, much of the comment friendly to the agency
has stressed its contribution to a new synthesis, one which would unite
positive government the welfare or service state with a rigorous ad-
herence to the principles of democracy.
What is this organization which is thought to embody an ideal so
eagerly sought ? What is the nature of this democratic technique ? What
are its implications and consequences? What will a close and critical
study of the organization in action tell us about these problems? These
questions have yet to be satisfactorily answered. To seek a partial an-
swer, a study was undertaken, during 1942-1943, with attention focused
primarily upon the Authority's "democratic" or "grass roots" method.
This inquiry was based upon the assumption that no prior personal com-
mitment to the TVA as a political symbol ought to interfere with a
realistic examination. It was an inquiry which did not hesitate to seek
out informal and unofficial sources of information. And it began with
certain ideas about the nature of the administrative process which seem
helpful in uncovering the underlying forces shaping leadership and
policy.
The Tennessee Valley Authority was created by Congress in May,
1933, as a response to a long period of pressure for the disposition of
government-owned properties at Muscle Shoals, Alabama. During the
First World War, two nitrate plants and what was later known as Wil-
son Dam were constructed, at a cost of over $100,000,000. For the next
fifteen years, final decision as to the future of these installations hung
fire. The focal points of contention related to the production and dis-
tribution of fertilizer and electric power, and to the principle of gov-
ernment versus private ownership. Two presidential commissions and
protracted congressional inquiries recorded the long debate. At last,
with the advent of the Roosevelt administration in 1933, the government
assumed responsibility for a general resolution of the major issues.
The TVA Act as finally approved was a major victory for those who
favored the principle of government operation. The Muscle Shoals in-
vestment was to remain in public ownership, and this initial project was
to be provided with new goals and to be vastly extended. A great public
power project was envisioned, mobilizing the "by-product" of dams
built for the purpose of flood control and navigation improvement on
the Tennessee River and its tributaries. Control and operation of the
nitrate properties, to be used for fertilizer production, was also author-
ized, although this aspect was subordinated in importance to electricity.
These major powers authority to construct dams, deepen the river
Selznick : T VA and the Grass Roots 5
channel, produce and distribute electricity and fertilizer were dele-
gated by Congress to a corporation administered by a three-man board
of directors.
If this had been all, the project would still have represented an im-
portant extension of government activity and responsibility. But what
began as, and what was generally understood to be, primarily the solu-
tion of a problem of fertilizer and power emerged as an institution of
far broader meaning. A new regional concept the river basin as an
integral unit was given effect, so that a government agency was created
which had a special responsibility neither national nor state-wide in
scope. This offered a new dimension for the consideration of the role of
government in the evolving federal system. At the same time, the very
form of the agency established under the Act was a new departure. There
was created a relatively autonomous public corporation free in impor-
tant aspects from the normal financial and administrative controls exer-
cised over federal organs. Further, and in one sense most important, a
broad vision of regional resource development in a word, planning
informed the conception, if not the actual powers, of the new organi-
zation.
The Message of the President requesting the TVA legislation did
much to outline that perception : "It is clear," wrote Mr. Roosevelt, "that
the Muscle Shoals development is but a small part of the potential public
usefulness of the entire Tennessee River. Such use, if envisioned in its
entirety, transcends mere power development : it enters the wide fields of
flood control, soil erosion, afforestation, elimination from agricultural
use of marginal lands, and distribution and diversification of industry.
In short, this power development of war days leads logically to national
planning for a complete river watershed involving many States and the
future lives and welfare of millions. It touches and gives life to all forms
of human concerns." To carry out this conception, the President recom-
mended "legislation to create a Tennessee Valley Authority a corpora-
tion clothed with the power of government but possessed of the flexibility
and initiative of private enterprise. It should be charged with the broad-
est duty of planning for the proper use, conservation, and development
of the natural resources of the Tennessee River drainage basin and its
adjoining territory for the general social and economic welfare of the
Nation."
This special regional focus and broad scope of the project have given
it a character which reflects one of the major motifs of our time : the
need for some sort of integral planning, especially in key problem areas.
6 University of California Publications in Culture and Society
It is that character which has been caught up as a model for similar
projects in other areas. For the uniqueness of TVA is not that it is a
government-owned power business or conservation agency, but that it
was given some responsibility for the unified development of the re-
sources of a region.
Yet it must be said that although the agency and its program have
symbolized concentrated effort and planning, in fact the TVA has had
little direct authority to engage in large-scale regional planning. The
powers delegated to it were for the most part specific in nature, related
to the primary problems of flood control, navigation, fertilizer, and
power. In addition, authority to conduct studies and demonstrations of
a limited nature, but directed toward general welfare objectives, was
delegated to the President and by him to the Authority. This became
the basis for some general surveys and demonstration work in forestry,
local industrial development, community planning, and for work with
cooperatives.
More important, however, is that the Act permitted such discretion
in the execution of the primary purposes as would invite those in charge
to recognize the social consequences of specific activities such as the
effect upon farm populations and urban communities of the creation of
large reservoirs and to assume responsibility for them. This assump-
tion of responsibility invests the administration with an important
planning function, though it is indirect and remains modifiable as
circumstances may demand. In addition, there remained administrative
freedom to devise methods of dealing with local people and institutions
which would reflect the democratic process at work. Perhaps of equal
importance is that the idea of planning associated with TVA accords
this agency a central status in the consideration of the problems and
the future of the Tennessee Valley region.
In the light of this weak delegation of broad planning powers, and
the tendency of Congress to restrict developmental functions, it is prob-
able that the significance of TVA in relation to democratic planning
comes primarily from the infusion of specific tasks with a sense of social
responsibility. In the purchase of lands, in the distribution of fertilizer
and power, in personnel policy in those functions which are a neces-
sary part of the execution of its major and clearly delegated responsi-
bilities the TVA has normally taken account of the people of the area,
with a view to adjusting immediate urgencies to long-term social policy.
This, of course, is not the same as devising and executing a frontal plan
for the reconstruction of the economy or institutions of an area. And
Selznick: TV A and the Grass Roots 1
yet, whichever view is emphasized whether one conceives of TVA's
limited regional planning as a portent of fuller ventures along that line,
or whether one thinks of planning as simply an adjunct of specific re-
sponsibilities we have something to learn from a study of the organi-
zation itself and of the methods developed in the execution of its tasks.
"Organization" and "method" are key words. Wherever we turn in
considering the implications of a program for democracy these terms
are inevitably involved. No democratic program can be unconcerned
about the objectives of a course of action, especially as they affect popu-
lar welfare. But the crucial question for democracy is not what to strive
for, but by what means to strive. And the question of means is one of
what to do now and what to do next and these are basic questions in
politics.
If the problem of means is vital, it is also the most readily forgotten.
"Results," "achievement," and "success" are heady words. They induce
submission and consent, thus summoning rewards for diligence and
labor and they also enfeeble the intellect. For the results which most
readily capture the imagination are external, colorful, concrete. They
are the stated goals of action. Their achievement lends reality, whole-
someness, and stature to the enterprise as a whole.
But methods are more elusive. They have a corollary and incidental
status. A viable enterprise is sustained in the public eye by its goals, not
its methods. Means are variable and expedient. Their history is forgotten
or excused. Here again the concrete and colorful win easiest attention.
Where incorrect methods leave a visible residue a rubbled city or
wasted countryside, then methods may gain notice. But those means
which have long-run implications for cultural values, such as democracy,
are readily and extensively ignored.
When we speak of methods, we speak in the same breath of instru-
ments. Policies, decisions as to "how to proceed," require execution.
Execution in turn implies a technology. We are familiar with the kind
of technology which includes machines and tools of all sorts, handled
and manipulated in more or less obvious ways. We are even reasonably
familiar with the technology of economic and military organization,
geared to the achievement of technical objectives, qualified and informed
by the criteria of efficiency. But when we move into that area of tech-
nology which is related to the creation, defense, or reintegration of
values, such as democracy, we find ourselves less assured. Yet the signifi-
cance of this noneconomic technology, under the conditions of mass
society and cultural disintegration, is of primary importance for what-
8 University of California Publications in Culture and Society
ever we may wish to do about that vague but demanding reality which
we call our "way of life." Propaganda agencies, mass parties, unions,
educational systems, churches, and governmental structures have a com-
mon aspect in that, more or less directly, they work upon and seriously
affect the evolving values, the spirit, of contemporary society. Further-
more, there is a growing tendency for this effect to be conscious, to be-
come an ordered technology available to those who have a stake in
changing sentiment or social policy.
One of the pervasive obstacles to the understanding and even the
inspection of this technology is ideology or official doctrine. By the very
nature of their function, all those forces which are concerned about the
evolution of value-impregnated methods, or public opinion itself, have
a formal program, a set of ideas for public consumption. These ideas
provide a view of the stated goals of the various organizations political
or industrial democracy, or decentralization, or the like as well as of
the methods which are deemed crucial for the achievement of these goals.
It is naturally considered desirable for the attention of observers to be
directed toward these avowed ideas, so that they may receive a view of
the enterprise consistent with the conception of its leadership. All this
in the often sincere conviction that precisely this view is in accord with
the realities of the situation and best conveys the meaning and signifi-
cance of the project under inspection.
However much we may be impressed by what a group says about its
methods or its work, there is adequate justification for uneasiness and
doubt. This doubt has its source in our general understanding of the
persistent tendency for words to outrun deeds, for official statement
and doctrine to raise a halo over the events and activities themselves.
That this is a natural disposition among responsible men is well under-
stood, and a gap of some sort between the idea and the act is normally
expected. But what is less well understood, or at least less generally
applied to objects of public esteem, is the tendency for ideas to reflect
something more than enthusiasm or more or less pardonable pride. The
functions of a doctrine may be more subtle and more significant, related
to the urgent needs of leadership and to the security of the organization
itself. Such functions, when relevant, cast a deeper shadow and indicate
the need for more searching questions. In Part I of this study, we have
critically analyzed TVA's official doctrine in relation to democratic
planning the policy of grass-roots administration as a contribution to
democracy. The analysis points to underlying issues and problems not
directly evident when we speak of the normal and anticipated gap be-
tween avowed statement and actual practice.
Selznick : T VA and the Grass Roots 9
Though official statements and theories are important, an undue con-
centration upon what men say diverts attention from what they do. This
is especially true with respect to the methods utilized in the execution
of a program, for these are particularly difficult to view realistically. It is
often sufficiently troublesome to attain a clear picture of the formal,
stated methods in use, without pressing inquiry as to the less obvious
but vital informal behavior of key participants. Yet it is precisely into
the realm of actual behavior and its significance for evolving structures
and values that we must move if this kind of inquiry is to realize its
possibilities.
The instruments of planning are vitally relevant to the nature of the
democratic process. The TVA is many things, but most significant for
our purposes is its status as a social instrument. It is this role as in-
strument with which this study is directly concerned. Or, to emphasize
another word, it is TVA as an organization to which our attention is
directed. Thus it is not dams or reservoirs or power houses or fertilizer
as such, but the nature of the Authority as an ordered group of working
individuals, as a living institution, which is under scrutiny.
In searching out organizational behavior and problems as keys to
understanding the implications of TVA for democratic planning, we
are entering a field of inquiry which probes at the heart of the demo-
cratic dilemma. If democracy as a method of social action has any single
problem, it is that of enforcing the responsibility of leadership or
bureaucracy. A faith in majorities does not eliminate the necessity for
governance by individuals and small groups. Wherever there is or-
ganization, whether formally democratic or not, there is a split between
the leader and the led, between the agent and the initiator. The phenom-
enon of abdication to bureaucratic directorates in corporations, in trade
unions, in parties, and in cooperatives is so widespread that it indicates
a fundamental weakness of democracy. For this trend has the conse-
quence of thrusting issues theoretically decided by a polity into the
field of bureaucratic decision.
The term "bureaucracy" has an invidious connotation, signifying
arbitrary power, impersonality, red tape. But if we recognize that all
administrative officials are bureaucrats, the bishop no less than the tax
collector, then we may be able to understand the general nature of the
problem, separating it from the personal qualities or motives of the
individuals involved. Officials, like other individuals, must take heed
of the conditions of their existence. Those conditions are, for officials,
organizational : in attempting to exercise some control over their own
10 University of California Publications in Culture and Society
work and future they are offered the opportunity of manipulating per-
sonnel, funds, and symbols. Among the many varied consequences of
this manipulation, the phenomena of inefficiency and arbitrariness are
ultimately among the least significant. The difference between officials
and ordinary members of an organized group is that the former have
a special access to and power over the machinery of the organization ;
while those outside the bureaucratic ranks lack that access and power.
If we are to comprehend these bureaucratic machines, which must
play an indispensable role in any planning venture, it is essential to
think of an organization as a dynamic conditioning field which effectively
shapes the behavior of those who are attempting to remain at the helm.
We can best understand the behavior of officials when we are able to
trace that behavior to the needs and structure of the organization as a
living social institution.
The important point about organizations is that, though they are tools,
each nevertheless has a life of its own. Though formally subordinated
to some outside authority, they universally resist complete control. The
use of organizational instrumentalities is always to some degree pre-
carious, for it is virtually impossible to enforce automatic response to
the desires or commands of those who must employ them. This general
recalcitrance is recognized by all who participate in the organizational
process. It is this recalcitrance, with its corollary instability, which is
in large measure responsible for the enormous amount of continuous
attention which organizational machinery requires. There are good rea-
sons, readily grasped, for this phenomenon.
The internal life of any organization tends to become, but never
achieves, a closed system. There are certain needs generated by organiza-
tion itself which command the attention and energies of leading partici-
pants. The moment an organization is begun, problems arise from the
need for some continuity of policy and leadership, for a homogeneous
outlook, for the achievement of continuous consent and participation on
the part of the ranks. These and other needs create an intricate system
of relationships and activities, formal and informal, which have pri-
marily an internal relevance. Thus leadership is necessarily turned in
upon itself. But at the same time, no organization subsists in a vacuum.
Large or small, it must pay some heed to the consequences of its own
activities (and even existence) for other groups and forces in the com-
munity. These forces will insist upon an accounting, and may in self-
defense demand a share in the determination of policy. Because of
this outside pressure, from many varied sources, the attention of any
Selznick: TVA and the Grass Roots 11
bureaucracy must be turned outward, in defending the organization
against possible encroachment or attack.
These general considerations, which have been stated here in a sum-
mary way, should lead to a more discerning study of any administrative
agency. They direct us (1) to seek the underlying implications of the
official doctrine of the agency, if it has one; (2) to avoid restriction to
the formal structure of the organization, as that may be outlined in
statutes, administrative directives, and organization charts; and (3) to
observe the interaction of the agency with other institutions in its area
of operation. Throughout, a search for the internally relevant in or-
ganizational behavior, especially that which is related to self -defensive
needs, is a primary tool of such analysis. 1
It will probably bear emphasis that the significance of TVA for
democratic planning lies not so much in its program, or in its accom-
plishments, as_in_its L mej;hodsjmd in its n^u^^s_anjor^amzaj;ion^Even
though its planning powers are limited, the TVA does represent an
experiment, an adventure in executing broad social responsibilities for
the development of a unified area. Furthermore, its type of organization
is proffered as a model for governmental planning in other areas. This
point has been clearly recognized within TVA itself :
Few of the activities of TVA are unique as public responsibilities. The Government
of the United States has been constructing waterways and building works for flood
control for more than a century. State and Federal agencies have engaged in technical
research, and surveys of mineral and forestry resources have been carried on with
public funds for many years. The TVA is not the first instance in which the Federal
Government has sold electric power. Aid to and stimulation of business opportunities
in industrial development, employment, farming, and other fields has become a
familiar role of Government, State and National.
1 It appears that this institutional approach (not, of course, original with the
author) to the study of administrative organization may be the avenue to an enlarge-
ment of the horizon of inquiry in this field. In a sense, this approach and this study
are a response to such criticism as that voiced by Donald Morrison in his review of
the series, Case Reports in Public Administration: "To put the matter succinctly,
the subject-matter of public administration has been defined so as to leave a no-man's
land of significant problems, flanked on one side by the students of administration
and on the other by political theorists. The problems thus isolated have their origin
in the fact that in its fundamental aspect administration is governance One such
problem, perhaps the most urgent, is to develop and strengthen ways of insuring that
government by the bureaucracy does not destroy the democratic pattern of our society.
Unless it is assumed that such insurance lies in the perfection of organizational struc-
ture and techniques of fiscal and personnel management, the present series of case
studies does not deal with this matter. Many persons believe that the TVA experiment
is suggestive of ways of democratizing bureaucratic government. Ten TVA studies
are published in Case Eeports, but none deals with the integration of the TVA pro-
gram into the social and economic life of the area" (Public Administration Review,
V,l [Winter, 1945], 85).
12 University of California Publications in Culture and Society
It is in the integration and the correlation on a regional basis of these various
activities under a single, unified management that the Tennessee Valley Authority
represents a pioneer undertaking of government. For the first time a President and
Congress created an agency which was directed to view the problems of a region as a
whole. 2
If the power granted to the Authority was not sufficient fully to execute
that broad responsibility, still the vision has remained. It is the con-
ception of an administrative instrument created to fulfill necessary
planning functions within the framework of democratic values.
If TVA as instrument is the focus of attention, and if we are prepared
to think of the Authority as a living social organization, we may expect
that in one way or another the Authority will have been caught up in
and shaped by its institutional environment. This expectation becomes
especially relevant as we note (1) the TVA's official avowal of a special
democratic relation to certain local institutions "close to the people," a
doctrine which will be discussed in detail below ; and (2) that TVA did
not arise out of the expressed desires of the local area, and consequently
was faced with a special problem of adjustment. Bach of these points
lends weight to the anticipation that in the Authority's relation to its
own grass roots we may find significant material of general interest to
those who wish to learn the lessons of the TVA experience.
Given such an anticipation, the problem for this inquiry became one
of finding a significant vantage point from which to examine this grass-
roots relationship. The question thus posed required some sort of theory,
a set of ideas which could point a way to the most vital aspects of the
situation. The theory which seemed to make sense in the light of a general
understanding of the materials was so formulated as to bring together
in a single over-all analysis (1) the avowed contribution of TVA to
democratic planning, through a grass-roots method of executing its
responsibilities; (2) the self -defensive behavior of the organization as
it faced the need to adjust itself to the institutions of its area of opera-
tion ; (3) the consequences for policy and action which must follow upon
any attempt to adjust an organization to local centers of interest and
power. Put in a few words, this involved the hypothesis that the Author-
ity's grass-roots policy as doctrine and as action must be understood as
related to the need of the organization to come to terms with certain local
and national interests; and that in actual practice this procedure re-
sulted in commitments which had restrictive consequences for the policy
and behavior of the Authority itself.
2 "The Widening of Economic Opportunity through TVA," pamphlet adapted from
an address by David E. Lilienthal, Director, TVA, at Columbia University, New
York, N.Y., January, 1940 (Washington: Government Printing Office, 1940), p. 15.
Selznick : T VA and the Grass Roots 13
In order to handle this problem most effectively, it has been found
necessary to introduce a concept which, while not new, is somewhat
unfamiliar. This is the idea of cooptation 3 often the realistic core of
avowedly democratic procedures. To risk a definition: cooptation is
the process of absorbing new elements into the leadership or policy-
determining structure of an organization as a means of averting threats
to its stability or existence. With the help of this concept, we are enabled
more closely and more rigorously to specify the relation between TVA
and some important local institutions and thus uncover an important
aspect of the real meaning and significance of the Authority's grass-roots
policy. At the same time, it is clear that the idea of cooptation plunges
us into the field of bureaucratic behavior as that is related to such demo-
cratic ideals as "local participation."
Cooptation tells us something about the process by which an institu-
tional environment impinges itself upon an organization and effects
changes in its leadership, structure, or policy. Cooptation may be formal
or informal, depending upon the specific problem to be solved.
Formal cooptation. When there is a need for the organization to
publicly absorb new elements, we shall speak of formal cooptation. This
involves the establishment of openly avowed and formally ordered rela-
tionships. Appointments to official posts are made, contracts are signed,
new organizations are established all signifying participation in the
process of decision and administration. There are two general conditions
which lead an organization to resort to formal cooptation, though they
are closely related :
1. When the legitimacy of the authority of a governing group or
agency is called into question. Every group or organization which at-
tempts to exercise control must also attempt to win the consent of the
governed. Coercion may be utilized at strategic points, but it is not
effective as an enduring instrument. One means of winning consent is
to coopt into the leadership or organization elements which in some way
reflect the sentiment or possess the confidence of the relevant public or
mass and which will lend respectability or legitimacy to the organs of
control and thus reestablish the stability of formal authority. This de-
vice is widely used, and in many different contexts. It is met in colonial
countries, where the organs of alien control reaffirm their legitimacy by
coop ting native leaders into the colonial administration. We find it in
3 With some modifications, the following statement of the concept of cooptation
is a repetition of that presented in the author's "Foundations of the Theory of Or-
ganization," American Sociological Review, XIII, 1 (February, 1948), pp. 33-35. For
further discussion of cooptation see below, pp. 259261.
14 University of California Publications in Culture and Society
the phenomenon of "crisis-patriotism" wherein normally disfranchised
groups are temporarily given representation in the councils of govern-
ment in order to win their solidarity in a time of national stress. Coopta-
tion has been considered by the United States Army in its study of
proposals to give enlisted personnel representation in the courts-martial
machinery a clearly adaptive response to stresses made explicit during
World War II. The "unity" parties of totalitarian states are another
form of cooptation ; company unions or some employee representation
plans in industry are still another. In each of these examples, the re-
sponse of formal authority (private or public, in a large organization
or a small one) is an attempt to correct a state of imbalance by formal
measures. It will be noted, moreover, that what is shared is the responsi-
bility for power rather than power itself.
2. When the need to invite participation is essentially administrative,
that is, when the requirements of ordering the activities of a large
organization or state make it advisable to establish the forms of self-
government. The problem here is not one of decentralizing decision but
rather of establishing orderly and reliable mechanisms for reaching a
client public or citizenry. This is the "constructive" function of trade
unions in great industries where the unions become effective instruments
for the elimination of absenteeism or the attainment of other efficiency
objectives. This is the function of self-government committees in housing
projects or concentration camps, as they become reliable channels for
the transmission of managerial directives. Usually, such devices also
function to share responsibility and thus to bolster the legitimacy of
established authority. Thus any given act of formal cooptation will tend
to fulfill both the political function of defending legitimacy and the
administrative function of establishing reliable channels for communi-
cation and direction.
In general, the use of formal cooptation by a leadership does not
envision the transfer of actual power. The forms of participation are
emphasized but action is channeled so as to fulfill the administrative
functions while preserving the locus of significant decision in the hands
of the initiating group. The concept of formal cooptation will be utilized
primarily in the analysis of TVA's relation to the voluntary associa-
tions established to gain local participation in the administration of the
Authority's programs.
Informal cooptation. Cooptation may be, however, a response to the
pressure of specific centers of power within the community. This is not
primarily a matter of the sense of legitimacy or of a general and diffuse
Selznick : T VA and the Grass Roots 15
lack of confidence. Legitimacy and confidence may be well established
with relation to the general public, yet organized forces which are able
to threaten the formal authority may effectively shape its structure and
policy. The organization faced with its institutional environment, or the
leadership faced with its ranks, must take into account these outside
elements. They may be brought into the leadership or policy-determining
structure, may be given a place as a recognition of and concession to the
resources they can independently command. The representation of in-
terests through administrative constituencies is a typical example of this
process. Or, within an organization, individuals upon whom the group
is dependent for funds or other resources may insist upon and receive
a share in the determination of policy. This type of cooptation is typi-
cally expressed in informal terms, for the problem is not one of respond-
ing to a state of imbalance with respect to the "people as a whole" but
rather one of meeting the pressure of specific individuals or interest
groups which are in a position to enforce demands. The latter are inter-
ested in the substance of power and not necessarily in its forms. More-
over, an open acknowledgment of capitulation to specific interests may
itself undermine the sense of legitimacy of the formal authority within
the community. Consequently, there is a positive pressure to refrain
from explicit recognition of the relationship established. This concept
will be utilized in analyzing the underlying meaning of certain formal
methods of cooperation initated in line with the TVA's grass-roots policy.
Cooptation reflects a state of tension between formal authority and |
social power. This authority is always embodied in a particular structure !
aM leadership, but social power itself has to do with subjective and \
obj ecti ve factors which control the loyalties a nd potential manipulab ility
of the community. Where the formal authority or leadership reflects real
social power, its stability is assured. On the other hand, when it becomes
divorced from the sources of social power its continued existence is
threatened. This threat may arise from the sheer alienation of sentiment
or because other leaderships control the sources of social power. Where a
leadership has been accustomed to the assumption that its constituents
respond to it as individuals, there may be a rude awakening when
organization of those constituents creates nucleuses of strength which
are able to effectively demand a sharing of power.
The significance of cooptation for organizational analysis is not simply
that there is a change in or a broadening of leadership, and that this is
an adaptive response, but also that this change is consequential for the
character and role of the organization or governing body. Cooptation
16 University of California Publications in Culture and Society
results in some constriction of the field of choice available to the organ-
ization or leadership in question. The character of the coopted elements
will necessarily shape the modes of action available to the group which
has won adaptation at the price of commitment to outside elements. In
other words, if it is true that the TVA has, whether as a defensive
or as an idealistic measure, absorbed local elements into its policy-
determining structure, we should expect to find that this process has had
an effect upon the evolving character of the Authority itself. From the
viewpoint of the initiators of the project, and of its public supporters,
the force and direction of this effect may be completely unanticipated.
The important consideration is that the T VA's choice of methods could
not be expected to be free of the normal dilemmas of action. If the senti-
ment of the people (or its organized expression) is conservative, demo-
cratic forms may require a blunting of social purpose. A perception of
the details of this tendency is all important for the attempt to bind
together planning and democracy. Planning is always positive for the
fulfillment of some program, but democracy may negate its execution.
This dilemma requires an understanding of the possible unanticipated
consequences which may ensue when positive social policy is coupled
with a commitment to democratic procedure. The description and analy-
sis which follows, in tracing the consequences of TVA's grass-roots policy
for the role and character of the organization, may cast some light upon
that problem. 4
* The notion of "unanticipated consequence" referred to in this section is central
to this study. See below, pp. 253-259, for a theoretical statement of the problem.
Part One
OFFICIAL DOCTRINE PARAPHRASE
AND ANALYSIS
CHAPTER I
THE IDEA OF A " GRASS-ROOTS"
ADMINISTRATION
"Administration" means more than organization charts. Go in one administrative
direction and you have loss of liberty; go far enough and you have decisions enforced
by the Gestapo and the lash. Go in the other direction and you have people partici-
pating in the decisions of their government actively and with considerable zeal, an
increase in freedom and the corresponding increase in responsibility and discipline.
Proceeding on this basis, I began a series of public statements on organizational
characteristics of the TVA, particularly decentralization of administration as a
method of securing the participation of the people of the Valley in the TVA under-
taking. 1 DAVID E. LlLIENTHAL
THE TENNESSEE VALLEY AUTHORITY has been the subject of widespread
comment and study in all parts of the world. In Central Europe, in the
Philippines, in Palestine, in China, wherever, indeed, new methods of
approach to the problems of resource development and social planning
have been discussed, the TVA idea has been in the forefront. TVA has
become not merely an administrative model and prototype, but a symbol
of the positive, benevolent intervention of government for the general
welfare.
In America, too, the TVA is unquestionably a rallying-point for those
who favor a welfare state. The defense of the TVA and the extension
of its methods are accepted among these groups as an elementary duty.
What is known in the United States as the progressive movement (essen-
tially the forces which comprised the popular base of the New Deal)
has treated the TVA as a symbol of its aspirations and has been quick
to muster its forces behind the agency and the ideas it symbolizes when-
ever occasion demanded. This has been evident in the unequivocal de-
fense of the TVA organization itself in controversies over finances and
accountability ; and especially in the vigorous espousal of the TVA idea
as a model for regional development in other areas of the United States 1 .
For these forces, support of the TVA is a ready criterion of political
acceptability.
It is primarily as a symbol that TVA excites allegiances and de-
nunciation. In its capacity as symbol, the organization derives meaning
a mimeographed transcript of remarks by David E. Lilienthal to the Con-
ference on Science, Philosophy, and Eeligion, Columbia University, New York
(August 28, 1942).
[19]
20 University of California Publications in Culture and Society
and significance from the interpretations which others place upon it.
The halo thus eagerly proffered is in large measure a reflection of the
needs and problems of the larger groups which require the symbol and
use it. Hence controversy over the TVA may proceed irrespective of a
close examination of the organization itself. When examinations are
undertaken, they are made to exalt TVA or condemn it : on the one
hand, as positive government liberating and advancing the people whom
it serves, or, on the other, as a nefarious encroachment on free enter-
prise, with subversive intent. This is not surprising, for the symbol is
necessarily caught up in and manipulated by the broader issues and
forces involved in social conflict. Since it is in the context of the larger
struggle that basic political decisions are made, it is to be expected that
the TVA should there lose its special identity and become merely an
instrument and a focal point of attack and defense.
Yet however significant its symbolic feature may be, TVA is and must
be more than an idea. It is a living organization in a concrete social
environment. From that obvious statement many things follow. The
special leadership of such an organization is divided from the general
leadership of its diffuse support by the quality of responsibility. To
those for whom the idea is primary, the symbol is enough and the actual
organization which embodies it at any particular time somewhat irrele-
vant. But the symbol become apparatus generates its own problems for
whose solution the administrative leadership is directly responsible.
These new problems are technical, and go beyond political loyalty to
an objective ; they arise out of the need to weigh means and judge con-
sequences in the context of practical action. The administrative leader-
ship is turned in upon itself, preoccupied with the tools at hand and
with the concrete choices which must be made to implement the general
policy it seeks to execute. Whatever the ultimate national implications
of its choices, the organization in action must deal primarily with those
who are immediately and directly affected by its intervention. In order
to endure, it cannot depend only or even primarily upon the diffuse sup-
port of elements not directly involved in its work; its administrative
leadership must find support among local institutions, and develop
smooth working relationships with them. It must avoid a continuous
atmosphere of crisis and conflict which may lead in the first instance to
disorganization and frustration and in the long run to the upcropping
of significant threats to the very existence of the organization itself. In
short, the institution must seek some sort of equilibrium with the en-
vironment in which it lives.
SelznicJc: TVA and the Grass Roots 21
To point to such an adjustment, or the need for it, is to speak of the
normal course of events. Experienced participants in the organizational
process are familiar with the continuous striving for adjustment, though
few have spoken self-consciously of it. The management of its details is
accepted as part of the ordinary common sense of administrative leader-
ship. But it is often observable that where institutions have symbolic
meaning beyond (and often irrelevant to) their own structure and be-
havior, a blindness to the less rosy aspects of organizational life arises.
Explicit statements which trace the history of the organization in terms
of compromise and mediation are rejected out of hand, or at best, when
accepted, are deemed shocking exposures.
If we understand that the leaders of an operating organization have
problems separate from those who function primarily in the realm of
symbols, then it should be of particular interest to examine the ideas
and theories of that administrative leadership. Such an examination may
reveal meanings evolving from the problems of the organization as such,
and bearing implications not altogether consonant with the total view
of those who accept and support the symbol-at-large.
The TVA is a particularly valuable subject for study in these terms.
For its leaders have been especially active more so than most other
governmental agencies, though there are few which do not venture at all
upon this road in propagating a systematic formulation of its own
meaning and significance. This self -analysis of the role of TVA has been
elaborated at length and presented to the public at every possible oppor-
tunity. Far from being a casual or incidental analysis, it is presented as
a fundamental key to an understanding of the Authority. It has been
made a part of almost all important speeches and publications which
describe the agency to its special publics and to the world. Inside the
organization, strenuous efforts are made to indoctrinate all who are in-
volved in administration on a policy-affecting level. The elaboration of
this interpretation has been, indeed, a well-developed and effective exer-
cise in administrative self -consciousness.
This chapter will be devoted to a statement of that doctrine. What
follows, therefore, represents a paraphrase of official TVA doctrine, as
that has been formulated verbally and in various documents, some pub-
lished, others not. An attempt is made to bring together most of the
arguments presented by TVA officials in support of the grass-roots
policy, and thus to present a rounded formulation of the TVA viewpoint.
This seems especially desirable because of the critical analysis which is
detailed in succeeding chapters. Hence, although responsibility for the
22 University of California Publications in Culture and Society
accuracy with which T VA views have been reflected in this chapter must
be accepted by the author, it should be understood that the intention is
to present the ideas of the TVA administration and not his own.
THE NEED FOR DECENTRALIZATION
OF ADMINISTRATION
David E. Lilienthal, until 1946 Chairman of the Board of Directors of
the TVA, has taken the leading role in public exposition of the official
TVA doctrine of grass-roots administration. While presented by him
partly in personal terms, the idea is nevertheless organically related to
the Authority as a whole. Lilienthal has specifically stated that the
fundamentals of the theory must be credited to his colleague on the
Board, Dr. Harcourt A. Morgan, former President of the University of
Tennessee. Moreover, within the organization, at in-service training
conferences, administrative seminars, and in numerous memorandums
on tangential subjects, the theory is repeatedly discussed as basic or-
ganization policy. Its significance therefore extends far beyond the
individual views of Lilienthal, though he may have provided a personal
coloration for bare essentials which might otherwise be differently, and
perhaps less forcefully or sympathetically, presented.
The context of the TVA thesis is the world-wide trend toward cen-
tralization and, concomitantly, the growing responsibilities of national
government. Centralization has been proceeding apace in all fields of
human organization. Efficiency has been, in this view, a rigorous leveler,
erasing the diversity of individual enterprise and local control in the
interests of large hierarchized units. This process operates not in gov-
ernment alone but in many other fields as well, and always brings with
it like and ambiguous consequences. In exchange for the benefits of order
and coordination, initiative has been stifled and the power of decision
indispensable element of democratic action lodged in far-off places,
remote from the beneficial influences of local areas which become merely
the objects of bureaucratic manipulation. Small businessmen, the inde-
pendent artisan, and farmer alike, have felt the enervating effect of the
concentration of economic and social control. By a similar logic, small
nations, too, have been unable to endure alone, and have reluctantly
found their way into some broader hegemony which provides security
in exchange for liberty.
In the wake of the general centralization of social (and especially
economic) life, has followed inevitably the centralization of public au-
thority. But this, like the centralization in other fields, is not unequivo-
Selznick: TVA and the Grass Roots 23
cally bad. Just as centrally managed private enterprises have achieved
lowered costs, more efficient and wider distribution, and the advancement
of science and invention, so too, centralized government has brought
improvements which cannot be denied out of hand. Above all, it is neces-
sary to see that in the ambiguous results of centralization there is a
problem, a dilemma, which must be recognized explicitly and boldly
faced so that new techniques of organization may be devised which, while
preserving the essential good, will eliminate the more critical evils. This>
problem is pressing, for the disappearance of small units and locaU
controls "lays bare the peculiar hazard of this modern world : the danger
implicit in vast size, the disaster consequent when power is exercised
far from those who feel the effect of that power, remote and alien to their
lives." 2
The recent history of American democracy has been in significant
part, according to TVA doctrine, a history of the simultaneous broaden-
ing of the responsibilities and the field of intervention of national
government. Although not without difficulties, and certainly without
complete uniformity, the executive, legislative, and judicial departments
of the federal government have alike accepted a widened view of the
fields of regulation and positive construction in which Washington agen-
cies might operate. These new national responsibilities have been ac-
cepted in response to (1) acute needs, such as unemployment, impossible
to ignore and yet beyond the power of the states to handle ; (2) the de-
mands of large interest groups, including labor and agriculture, organ-
ized on a national scale and viewing the power of the federal government
as both objective and instrument; (3) the growth of centralized industry
requiring the counterbalance of a federal government strong enough to
meet it on something like equal terms ; and (4) the growth of colleetivist
ideology, supporting and justifying the trend toward over-all integra-
tion. National problems have demanded national recognition, and doubt-
less will continue to do so. This trend is irreversible; it would be idle to
attempt to turn back the clock, to return to methods which offer no
answer to the real and urgent problems posed by modern society and
its technology.
At the same time, we are told, the question must be raised: is it
necessary that the exercise of federal functions be identified with top-
heavy organizations centered in and administered from Washington?
For the most part this identification has been made. As statutes have
2 David E. Lilienthal, "The TVA : A Step Toward Decentralization," address be-
fore the Graduate Faculty of Political and Social Sciences, New School for Social
Eesearch, New York City, April 3, 1940 (mim.).
24 University of California Publications in Culture and Society
been enacted recognizing new obligations of the national government,
their administration has been delegated to existing departments and
ad hoc agencies with headquarters in Washington, thus increasing the
staffs, responsibilities, and functions centered at the national capital. If
this procedure has sometimes effectively provided the services required,
it has done so while retaining and extending the basic conception of a
national government centrally administered.
This historic extension of centralized power has, in Lilienthal's view,
created a justified feeling of uneasiness and distrust. Ordinary people
and, increasingly, men in responsible posts in business and government
have been understandably fearful of an unchecked growth of a vast
administrative apparatus in Washington. It is in this proliferation of
Washington-oriented agencies rather than in the mere grant of power
to the federal authority that there is reason for fear. "This country is
too big for such a pyramiding of direct responsibilities. For in spite
of our triumphs over time and space, Washington is still remote from
the average citizen and is sheltered from participation in his daily
struggles." 3
An excessively centralized government is inherently disqualified, at
least in the United States, from fully promoting the welfare of its citi-
zens. This is a consequence of the significant differences in attitude and
custom which are the elements of diversity within our general cultural
and political unity. These differences cannot adequately be safeguarded
when programs are formulated purely in national terms. Methods of
approach to local people in such administrative contexts as employment
or land purchase may well be homogeneous if we consider only the in-
terests or convenience of the bureaucracy. But the people treasure their
special folkways, and protect them as basic elements of welfare and
security.
This fundamental evil of overcentralized government is accompanied
by another: the inhibition of action through the proliferation of red
tape. A centralized agency, remote from the field of operation, lays a
deadening hand upon its officers "on the line" by relieving them of the
responsibility for significant decision. In the interests of standardized
practice and accountability in detail, the national headquarters is driven
to insist that all important problems, and many that are not so impor-
tant, be referred up the hierarchy, "through channels," for action. The
3 David E. Lilienthal, "The TVA : An Experiment in the 'Grass Boots' Adminis-
tration of Federal Functions," address before the Southern Political Science Asso-
ciation, Knoxville, Tennessee, November 10, 1939 (Knoxville: TVA Information
Office, no date), p. 4.
Selznick: T VA and the Grass Roots 25
consequences of this procedure are well known : delays when action is
contemplated and the stifling of initiative in the face of a justified fear
of being pitted against the resistance of the bureaucracy.
From these observations Lilienthal derives a fundamental dilemma.
On the one hand, it must be conceded that increasingly large powers |
ought to be intrusted to the federal government, for there are too many
basic problems which cannot be handled through the organs of local
control ; on the other hand, the centralization of large powers is always
a menace to democracy. "It must be recognized that there is genuine
peril if the powers of federal government are hopelessly outdistanced
by the trend to centralized control in industry and commerce and fi-
nance"; still, "the dangers of centralized administration are all too
evident. They cannot be ignored."*
For his answer to this dilemma, Lilienthal has turned to (and widely
publicized) the formulations of an earlier commentator on American
democracy, Alexis de Tocqueville. De Tocqueville outlined a distinction
between centralized government and centralized administration, a dis-
tinction which has served as the theoretical anchor of the grass-roots
answer to the dilemma of centralization.
Centralization is a word in general and daily use, without any precise meaning
being attached to it. Nevertheless, there exist two distinct kinds of centralization,
which it is necessary to discriminate with accuracy.
Certain interests are common to all parts of a nation, such as the enactment of its
general laws and the maintenance of its foreign relations. Other interests are peculiar
to certain parts of the nation, such, for instance, as the business of the several town-
ships. When the power that directs the former or general interests is concentrated in
one place or in the same persons, it constitutes a centralized government. To con-
centrate, in like manner in one place the direction of the latter or local interests,
constitutes what may be termed a centralized administration. 5
Lilienthal has endorsed and often quoted the following statement of De
Tocqueville :
Indeed, I cannot conceive that a nation can live and prosper without a powerful
centralization of government. But I am of the opinion that a centralized administra-
tion is fit only to enervate the nations in which it exists, by incessantly diminishing
their local spirit. Although such an administration can bring together at a given
moment, on a given point, all the disposable resources of a people, it injures the
renewal of those resources. It may ensure a victory in the hour of strife, but it
gradually relaxes the sinews of strength. It may help admirably the transient great-
ness of a man, but not the durable prosperity of a nation. 9
*I~bid., p. 7.
5 Alexis de Tocqueville, Democracy in America (New York: Knopf, 1945), I: 86.
"Ibid., p. 87. Whether Lilienthal has been correct in resting his argument on De
Tocqueville's analysis is not here in question.
26 University of California Publications in Culture and Society
De Tocqueville's strictures set a serious and enduring problem, but of
course they provide no solution. Yet the distinction between centralized
government and centralized administration points the way, in Lilien-
thal's view, to a solution in modern terms. It is a way which appears to
offer an alternative to the acceptance of "topheavy, cumbersome, cen-
tralized administration as the price of granting necessary powers to our
federal government." The answer lies in a new solution to the way in
which the powers of the national government are administered.
There is, of course, no single remedy, but it is believed in TVA that
the theory of administrative decentralization, properly understood and
wisely applied, will serve to check the growing tendency toward exces-
sive centralization of federal administration. This is not a new idea, for
there has been for some time an increasing awareness in Washington
of the need for a greater measure of decentralization. But for the most
part we have seen no fundamentally new departure, for the attempt by
federal departments to set up regional offices and otherwise establish
important nucleuses of administration in the field has left the old struc-
ture of control intact. The reins are held by the central offices and their
various control divisions ; these in turn are subject to the manifold re-
strictions (such as laws regulating procurement and accountability)
which serve to crystallize and perpetuate the existing system of central-
ized control.
There has been, however, one federal agency which serves as an ex-
ample of and an experiment in the decentralization of federal functions.
This is the Tennessee Valley Authority, "the boldest and perhaps most
far-reaching effort of our times to decentralize the administration of
federal functions. If it succeeds, if its methods prove to be sound, we
shall have added strength to the administrative defenses which protect
the future of our beleaguered democracy." 7 The TVA is invested with the
authority of the national government, and derives its power from the
exercise by the federal government of its constitutional prerogatives as
interpreted in our time. At the same time, administration of these powers
is effectively decentralized.
When the TVA was established, the programmatic responsibilities of
the federal government for the control of the natural resources of water
and land were brought together and treated as a unit in order to deal
effectively with the watershed of the Tennessee River as an integrated
area, a single problem in resource development. Thus one agency, lo-
cated in the area of operation, represented the federal government in
7 Lilienthal, "The TVA : An Experiment . . . ," p. 10.
Selznick : T VA and the Grass Boats 27
relation to a complex local problem. The TVA was not appended to any
existing department, but was given freedom of action and a substantial
measure of autonomy, subject only to the direct control of the President
and Congress. The TVA Act, the agency's basic charter, was so framed
as to give the Authority the power to make its own decisions, and thereby
such flexibility as would make possible a maximum adjustment to local
conditions. This delegation of discretion is the heart of the idea of grass-
roots administration.
The broad significance which Lilienthal attributes to the TVA as a
new departure in public administration is made additionally clear by
his response to the challenge of James Burnham's widely noted book,
The Managerial Revolution. This book singled out the TVA as an ex-
ample of the social revolution, now encompassing the earth, which, as
Burnham claimed, is raising the managers, among whom are the admin-
istrators of great corporations, to the heights of social power. It asserted
further that this new class would in its turn become an exploiting group,
using specialized skill and social position for its own class benefit. Lilien-
thal agrees that the extension of public enterprise has been great, and
that "government which has undertaken a major responsibility for pro-
duction will of course depend increasingly upon men in public service
skilled in the art and science of management." 8 But he denies that this is
inevitably a threat to democracy. We can, he asserts, introduce counter-
measures which will offset the managerial trend. Among these measures
is administrative decentralization :
That we are in grave danger of putting these fetters upon our arms no one can
deny. But we do have a choice. We can choose, if we will, between an exploiting
managerial class and managers bound by the principles of public service and demo-
cratic methods. Indeed managerial technicians, clothed as they must be with the
great power of their skill and of society's reliance upon them, ought to assume leader-
ship in protecting American democratic society from exploitation by the managers
themselves. That leadership can be expressed, to take one example, by a constant
search by public administrators for methods of avoiding overcentralization in admin-
istration. Every sophisticated manager knows that tyranny and exploitation feed
upon excessive centralization of administration. He knows that overcentralized ad-
ministration dries up the well-springs of initiative, of energy, and of independence
in any organization and most of all in government institutions. It is overcentraliza-
tion that gives a "clique of headquarters courtiers" an opportunity to maneuver and
natter their way to power which they are not qualified by their abilities to exercise.
Absentee government is the quickest way to raise up the exploiting managerial class
that Mr. Burnham's book predicts with such confidence. But these prophecies need
8 David E. Lilienthal, "Management Eesponsible or Dominant?" Public Admin-
istration Review, I (Summer, 1941), 391.
28 University of California Publications in Culture and Society
not be fulfilled ; we do have a choice, for the hazards of managerial exploitation can
be diminished by skillful efforts in the direction of decentralized administration of
centralized authority. 9
Thus the social problem to which the grass-roots theory is directed is
of sweeping proportions. And though the technique of federal decen-
tralization is not presented monolithically, as an only answer or a
panacea, still it is clear that, in the mind of its chief proponent, it is
more than a modest device linked to special problems within the federal
system.
MINIMUM ESSENTIALS
In a decade of existence as a decentralized federal agency, the TVA had
the opportunity to formulate the special requirements which differen-
tiate actual decentralization from the kind which represents little more
than lip service. The goals have been variously stated, but uniformly
include the following elements :
1. The responsible agency in the area of operation is permitted the
freedom to make significant decisions on its own account. This assumes
that the organization is to be independent, that it will not be made a part
of some larger administrative agency having the power to write rules
and regulations signed by the top administrator in Washington and
hence immediately applicable with the force of law to the local operating
organization. It further assumes that the field officers will be of a capacity
and standing which will permit them to exercise broad discretion in
adapting general policies to particular local situations, and to do this
in such a way that their authority in local matters will be recognized by
the local public. Moreover, this independent agency, if it engages in
business functions, is to be permitted wide corporate freedom in the
exercise of such managerial responsibilities as the selection of personnel,
procurement, and disposition of operating funds.
2. There must be active participation by the people themselves in the
programs of the public enterprise. The services of state and local agencies
are to be utilized, with the federal government providing leadership
that will strengthen rather than weaken or eliminate the existing agen-
cies. Management should devise means to enlist the active and conscious
participation of the people through existing private associations as well
as through ad hoc voluntary associations established in connection with
the administration of the agency's program.
3. The decentralized administrative agency is given a key role in co-
ordinating the work of state, local, and federal programs in its area of
u Loc cit.
Selznick: TV A and the Grass Roots 29
operation; and a regional development agency should le given primary
responsibility to deal with the resources of the area as a unified whole.
The place for the coordination of programs is in the field, away from
the top offices which are preoccupied with jurisdictional disputes and
organizational self-preservation. Coordination should be oriented to the
job to be done, centering federal authority and its administrative skills
and power upon the special needs and problems of the area.
These minimum essentials may be more clearly understood if they are
examined as (1) the concept of managerial autonomy, (2) the partner-
ship of TVA and local government, and (3) the ideal of basing unity
^administration upon the natural unity of a region as an area of opera-
tion in resource development.
MANAGERIAL AUTONOMY
The TVA itself, as an autonomous administrative agency, "a corporation
clothed with the power of government but possessed of the flexibility
and initiative of a private enterprise," 10 embodies the first of the essen-
tials of decentralized administration : the freedom to make significant
decisions on its own account. This freedom from centralized control is
in turn a condition for the realization of the second essential, for the
decentralized agency must be able to make its own choices, on the basis
of its experience in the field, in order to maximize the "participation of
the people themselves."
To this end, the TVA has demanded managerial autonomy 11 in its
relation to the federal government. The Authority has insisted that if
effective decentralization is to be achieved the organization must be
permitted to retain real administrative powers unhampered by the ad-
ministrative controls ordinarily exercised over government departments.
Considered broadly, this demand has a wider significance than the estab-
lishment of the conditions of decentralized administration. It is also
intended as revolt against the conception of government as a necessary
evil and government officials as inherently tainted. In the eyes of some
TVA officials, the stringent controls over personnel and financial policy
normally exercised by civil service and budgetary agencies represent a
10 Message of the President, 10 April 1933. House Doc. 15, 73d Cong., 1st sess.
11 This study is concerned primarily with the TVA's relation to local institutions,
as expressed in the implementation of the grass-roots policy. The problem of man-
agerial autonomy as such concerning the relation of the Authority to the federal
government is not directly relevant. However, space is allowed here to the official
position on managerial autonomy because of TVA's insistence that precisely such
autonomy is indispensable to a proper execution of a grass-roots policy. It has there-
fore seemed desirable to discuss autonomy in detail, since this chapter as a whole is
devoted to an exposition of the TVA viewpoint.
30 University of California Publications in Culture and Society
cultural lag, still attuned to a governmental structure limited in scope
and essentially parasitic rather than playing a significant positive role in
social and economic life. The rationale of strong housekeeping controls
seems to be weakened rather than strengthened by the conditions of the
new state with its ever broadening functions. A new problem arises:
not one of restricting the powers and functions of government, but of
developing new techniques which will permit, within the framework of
positive government, the exercise of initiative and the kind of independ-
ence which can forestall the rise of centralized bureaucracy. Can the
benefits of anticorruptionist controls be obtained outside of the old ad-
ministrative devices? This question, in the opinion of the Authority's
leadership, can now be answered positively.
It is doubtful that the experience thus far accumulated is sufficient
to permit any exhaustive listing of the administrative freedoms neces-
sary for that degree of managerial autonomy required by administrative
decentralization and yet consistent with a reasonable measure of broad
control by the President and Congress. But it is possible to state three
such conditions of autonomy which are considered fundamental by TVA.
1. Freedom from control ~by the Civil Service Commission. The Au-
thority has been free of control by federal civil sendee laws from its
inception, and its continued exemption from the Ramspeck Act of 1940
for the general extension of civil service has underlined its unique posi-
tion. The TVA Act of 1933 provided for the establishment of an internal
civil service system under the control of the Authority, by stipulating,
in Section 6 :
In the appointment of officials and the selection of employees for such operation and
in the promotion of any such employees or officials no political test or qualification
shall be permitted or given consideration but all such appointments and promotions
shall be given and made on the basis of merit and efficiency. Any member of said
Board who is found by the President of the United States to be guilty of violation
of this section shall be removed from office by the President of the United States.
And any appointee of said Board who is found by the Board to be guilty of a viola-
tion of this section shall be removed from office by said Board.
It is a well-known feature of TVA's administration that this provision
was taken to heart and translated into an effective policy from the
beginning with the vigorous support of Arthur E. Morgan, the first
chairman of the Board. In recruitment, classification, and labor rela-
tions the Authority has been widely praised as a model of progressive
personnel policy. There is some doubt whether control by the Civil Serv-
ice Commission in the early days of the Authority would have permitted
Selznick : T VA and the Grass Roots 31
the kind of experimental approach which was instituted. Since the merit
system in the Authority is secured by its basic Act, and ten years of
experience have indicated the feasibility of such a system under auton-
omous control. TVA maintains that the solution it offers to the problem
of centralization upholds high standards of civil service as well as the
principle of managerial autonomy. In the Authority's view the mainte-
nance of such standards does not require the sharing of administrative
control over personnel policy with another agency, particularly one
located in Washington.
The special character of TVA's requirement of freedom from civil
service controls although not, to be sure, from the principle of merit
and efficiency was evident in the Authority's early history and was
derived not only from its attempt to develop an experimental personnel
program but also from its necessity to be free to choose a personnel
adaptable to its policies. This required flexibility in recruitment, un-
hampered by regulations which assume that all who reach a certain
grade on an examination are equally fit for a position. Of considerable
importance in this connection was that TVA had to recruit in a situation
of struggle. The Authority was subject to bitter political and legal at-
tack from its inception, and required local control over personnel re-
cruitment in order to avoid the employment of hostile men for key
administrative posts. This was one of the points made by Senator Norris
in opposing the extension of civil service to TVA under the Kamspeck
Bill:
Now, Mr. Chairman, here is another thing I have advocated, that I believe in.
It is something which hasn't been presented even by the TVA although I have talked
to members of the TVA Board about it. Without referring to anyone in particular,
I want to say that especially while a period of construction in TVA is going on, a
period which will extend at least three years more, we never should make it possible
for a spy of what I have denominated the power trust, through civil-service regula-
tions, to get into the service of the TVA. That is another reason why I contend they
have a present civil service far superior to that of the ordinary Government civil
service.
It would be a comparatively easy matter for a group of men or women to pass
civil-service regulations and examinations and get appointments in the TVA, where
they would have access to the confidential files and confidential communications and
discussions that take place. For instance, while those great lawsuits were pending,
when millions in money would have been forthcoming in the twinkling of an eye to get
information on the confidential actions and communications that were going on within
the body of the TVA, there might have been great injury and sabotage. 12
12 Hearings before the Committee on Civil Service, U. S. Senate, on H. E. 960, "An
Act Extending the Classified Civil Service," 76th Cong., 3d sess. (1940), pp. 193-194.
32 University of California Publications in Culture and Society
The advantage of autonomous control over personnel policy, safeguarded
by the merit and efficiency section of the Act, lies in the ability of the
agency to choose the kind of men it wants, on the basis of an implicit
theory of a close relation between personnel and policy, and thus to
shape the character of the organization in a unified direction. The or-
ganization may then rely on the weight of personality and group atti-
tudes within its structure for the implementation of policy rather than
solely on administrative rules. Of course, all organizations will tend to
educate the administrative ranks after recruitment, with varying suc-
cess, in order to achieve a unified approach to questions of policy and
consequently a loyalty to the decisions of the leadership. But civil service
autonomy provides the opportunity of choosing personnel at the outset
on the basis of an estimate of the kind of men needed for the effectuation
of a given policy. It has the added advantage, not less important, of
permitting the flexible adjustment of personnel policies to the kinds of
administrative arrangements deemed especially applicable to the local
situation within which the agency must operate.
The relevance of the demand for managerial autonomy in civil service
matters to the Authority's policy of cooperation with local institutions
was recognized by the Joint Committee which investigated the TVA in
1938:
A further consideration of great weight is the relationship between the Authority
and various other organizations, public and private, elsewhere discussed in this report.
The Authority has been able to cooperate with other governmental agencies, Federal,
State, and local, sometimes by providing personnel or the means to be used in hiring
personnel. No evidence of corruption or inefficient use of this power has been sub-
mitted. It has served as a device for obtaining desirable results without arbitrary
use of Federal power and with a maximum of democratic participation. The Civil
Service regulations would have to be carefully limited to avoid serious interference
with this form of cooperation. 13
The Authority itself has emphasized the importance of an autonomous
civil service for a regionally decentralized administration. In opposing
the application of the Kamspeck Bill to its organization, the TVA pro-
tested that the character of the Authority demanded that "managerial
decisions should be made in the Valley, where the work was to be done,
not in Washington." 14
13 Eeport of the Joint Committee to Investigate the Tennessee Valley Authority,
Senate Doc. 56, 76th Cong., 1st sess. (1939), p. 60.
14 "Statement on Application of H. E. 960 to the Tennessee Valley Authority,"
Hearings before the Committee on Civil Service, U. S. Senate, 76th Cong., 3d sess.
(1940), p. 249.
Selznick : T VA and the Grass Roots 33
A large part of the work of the Authority is, moreover, carried on in cooperation
with State and local agencies and with other Federal agencies.
The Agricultural Extension Service, the county-agent system, the staffs and facil-
ities of State universities and departments, the municipal and county officials and
agencies throughout the Tennessee Valley, are all essential elements in the Author-
ity's program. It is through these agencies that the Authority sells its power, dis-
tributes its fertilizer, encourages improved land practices, promotes research in
industry and agriculture, contributes to health and related developmental programs.
The Authority is the only regional agency of its type in the country, and its relations
with other public agencies are therefore different in quality and importance from
those of any other Federal agency. Its staff must be specially qualified to carry on
a regional program through regional collaboration and participation. They are se-
lected not only for general administrative experience in organizing such cooperative
programs, but also for their special familiarity with regional problems, regional
procedures, and regional agencies.
There can be no genuine regional administration of Federal functions in the
Tennessee Valley if the responsibility for so important a phase of the Authority's
program as the selection of the men who carry on its work, and the determination
of the terms and conditions of their employment, is lifted from the discretion of
the board under the terms of its comprehensive statute and lodged in a separate
agency with headquarters in Washington. Such action would mean the beginning of
piecemeal disintegration of a regional administration. . . , 13
The final sentence of the above quotation is actually the key. For al-
though it might be shown that many of the concrete special problems
faced by TVA as a regional agency could be solved within the frame-
work of the federal civil service system, abdication of control over re-
cruitment would still be inacceptable to the Authority. It is the denial
of the principle of managerial autonomy implicit in the federal civil
service which is decisive, and the threat that (a) federal personnel con-
trol would be only the first of a series of inroads on the unique privileges
of the TVA within the federal system; or even if this were not true,
(b) a change in the policies of an outside Washington agency which
TVA cannot control might thus overturn existing arrangements in the
Valley or hinder further regional adaptation.
from control by the General Accounting Office. Under
the provisions of the Budget and Accounting Act of 1921, the Comp-
troller General of the United States an agent of the Congress in its
exercise of the power of the purse was given the responsibility of
assessing the legality of expenditures by federal departments. This
power is embodied in a procedure which makes it possible for the Comp-
troller General to disallow expenditures by refusing payment when a
particular transaction is deemed illegal.
15 Loc. cit.
34 University of California Publications in Culture and Society
From the viewpoint of managerial autonomy, control by the General
Accounting Office brings to bear upon the government agency the gen-
eral housekeeping laws used as criteria for the legality of the trans-
actions under scrutiny. Among examples of such statutes are "those
requiring that purchases of materials, with certain exceptions, be pre-
ceded by advertisement for competitive bids; that needed articles be
purchased from the federal prison industries if available from that
source ; that purchase of land on which is to be erected any public build-
ing be preceded by a favorable ruling of the Attorney General as to
the validity of the title ; that printing and binding be done in the Govern-
ment Printing Office ; that law books and periodicals not be purchased
with appropriated funds unless their purchase has been specifically
authorized by statute ; that no payments be made to any publicity expert
except from funds specifically appropriated for that purpose." 10
The Tennessee Valley Authority has claimed freedom from these stat-
utes, as well as from the authority of the Comptroller General to settle
and adjust its accounts. This freedom was challenged early in the history
of the Authority and resulted, after a considerable debate, in a clarifica-
tion of the TVA Act by Congressional amendment. This amendment,
enacted in November, 1941, placed the Authority definitely within the
jurisdiction of the Budget and Accounting Act of 1921, but at the same
time relieved it of any onerous consequences of that jurisdiction. The
Authority was authorized to make final settlements of claims and litiga-
tion, and the Comptroller General was restrained from withholding
funds "because of any expenditure which the Board shall determine to
have been necessary to carry out the provisions of said Act." 17
The TVA considers it fundamental to its status as a corporation en-
gaged in public business and therefore subject, at least in large part, to
judgment in financial terms, that it be permitted to make its decisions
on a business basis. This is especially important in such matters as
procurement and the settlement of litigation on an expedient basis. The
criterion of business efficiency does not of itself necessitate safeguarding
competition through advertisement for bids in procurement; but it is
concerned with obtaining materials of suitable quality at the lowest cost
and with the least delay. Thus the normal requirements of a business
enterprise appear to argue, in the TVA view, for granting managerial
autonomy to a decentralized regional authority with business functions.
16 David E. Lilienthal and Robert H. Marquis, "The Conduct of Business Enter-
prises by the Federal Government," Harvard Law Review, 54 (February, 1941), 578.
17 55 Stat. 775-776 (1941). The Authority's corporate freedom in this respect was
preserved under the provisions of the Government Corporation Control Act of 1945.
See 59 Stat. 597 (1945).
Selznick : T VA and the Grass Roots 35
3 Prfifidnft to ppply/revenues to current operational expenses.
Among the rights wMcw ie~l^ VSTEas defined as cracTaTlol^acEmmis-
trative existence has been that of freely disposing of its revenues from
power and other sources in the interests of efficient management. Moneys
received by the Authority are paid into a special fund in the Treasury,
and may be applied to operating expenses as determined by manage-
ment within the Authority. This, in addition to insuring continuity of
operations of its electric power plant, has permitted the Authority to
make budgetary determinations, such as those for travel expenses, sub-
stantially on its own account. The normal alternative in the federal de-
partments is that itemized budget allowances are made by Congress;
since moneys are available only from appropriations and for specified
items, these allowances are not only relatively inflexible but precarious.
It is the position of the TVA that its freedom from Congressional allo-
cation and appropriation permits management to make adjustments in
terms of needs as they arise and change, with an eye to over-all efficiency
rather than to the terms of the budget.
This entire procedure was called into question by Senator Kenneth
McKellar of Tennessee when he introduced a bill amending the TVA
Act so as to require its receipts to be paid into the Treasury on the same
basis as regular government departments. Although this amendment
was not enacted, it was conceived as a basic threat to the integrity of
the TVA organization, and provided the leadership with another oppor-
tunity to emphasize its demand for managerial autonomy. Lilienthal
put the matter strongly :
If the McKellar amendment were law, the TVA region would at a single blow be
pushed back industrially 25 years. Since continuity of power supply would depend
upon whether successive Congresses would appropriate funds from year to year, ex-
isting industries in the Tennessee Valley would be obliged to move out of the region
to a place where power supply could be depended upon without question. . . . Long-
term contracts have been made, in reliance upon the TVA's power to use its revenues
from those contracts to insure continuity of service. It would undermine those solemn
obligations to change the provisions of law which largely induced entry into those
contracts. I am not overstating the case when I say that the McKellar amendment
would make scraps of paper of more than 140 contracts made in good faith with
the U. S. A. 18
While the question of the continuity of power operations has been
stressed, it seems fair to say that equally important in the minds of the
TVA Directors has been the problem of maintaining the Authority free
18 Hearings on S. 2361, Committee on Agriculture and Forestry, U. S. Senate, 77th
Cong., 2d sess. (March, 1942), pp. 175-176.
36 University of California Publications in Culture and Society
from onerous political control. They have wished to avoid the introduc-
tion of ulterior considerations of political expediency into the determina-
tion of matters which should be decided on a business basis. Thus, in a
speech to an Alabama community at the time of the McKellar contro-
versy, Lilienthal said :
We meet here as typical representatives of several million of Americans engaged in
a unique undertaking. Our course in the future as in the past will be beset with
difficulties and opposition. Let us here resolve that we will fight at the drop of a
hat against any intrusion of predatory politics into this partnership. Let us continue
to administer this joint undertaking on principles of sound management. The people
of the communities which you represent join with communities in the other Valley
states to cry "hands off" to anyone who seeks to substitute politics in our partnership
for business management. 19
Viewed in this way it becomes clear that the issue cannot be resolved by
providing the Authority with a special fund to defray emergency ex-
penses while leaving all regular expenses, such as those for salaries,
travel, and information, subject to determination through the annual
appropriation procedure. For in general it is believed that political ex-
pediency may at any time become the deciding factor which will deprive
the agency of its needed funds quite apart from considerations of busi-
ness efficiency. With such a danger always present it would be inevitable
that the decisions of the agency itself would be made with a view to
minimizing possible threats to its funds, with the consequence that
management techniques would become tied to the problem of influencing
a Congressional committee. With political threats to its finances re-
moved, the TVA is free to make decisions on particular matters in terms
of internal efficiency judged by over-all results.
The three points stated above an independent personnel system,
freedom from housekeeping controls of the General Accounting Office,
and discretionary power to apply revenues to operating expenses have
been cited by TVA officials as fundamental principles of managerial
autonomy. Such autonomy represents one of the main features of the
grass-roots doctrine, having to do largely with the relations of a regional
agency to the federal administrative system.
Managerial autonomy, it will be noted, is not confined to the require-
ments of the Authority in its role as a business corporation. Thus, as
described above, the freedom from control by the Civil Service Commis-
19 David E. Lilienthal, "The Partnership of Federal Government and Local Com-
munities in the Tennessee Valley," address before gathering of community leaders of
northern Alabama, sponsored by Decatur Chamber of Commerce, Decatur, Ala., July
30,1942 (mim.).
SelznicJc : T VA and the Grass Boots 37
sion applies to the developmental aspects of the TV A, such as conserva-
tion and planning, as well as to its power business. The idea of business
efficiency as a criterion of professional management is thus extended
to traditional governmental functions.
THE PARTNERSHIP OF TYA AND THE PEOPLE'S
INSTITUTIONS
After some difficulties and initial disagreements, but still very early in
its history, the Authority denned its approach to cooperation with the
agencies and institutions already existing in the Valley. The alternatives
seemed to be two : either to take a line which assumed that the TVA
itself could and should carry out its programs by direct action; or to
accept as legitimate and efficient a method which would seek out and
even establish local institutions to mediate between the TVA and the
people of the area. It was felt that an imposed federal program would
be alien and unwanted, and ultimately accomplish little, unless it
brought together at the grass roots all the agencies concerned with and
essential to the development of a region's resources : the local commu-
nities, voluntary private organizations, state agencies, and cooperating
federal agencies. The vision of such a working partnership seemed to
define "grass-roots democracy at work."
In the Authority's view, the fundamental rationale of the partnership
approach is found in its implications for democracy. If the TVA can
be "shaped by intimate association with long-established institutions," 24
that will mean that its vitality is drawn from below. By working through
state and local agencies, the Authority will provide the people of the
Valley with more effective means by which to direct their own destinies.
The TVA may then become more integrally a part of the region, com-
mitted to its interests and cognizant of its needs, and thus removed in
thought and action from the remote impersonal bureaucracy of central-
ized government.
The moral dimension of the grass-roots approach has been emphasized
many times. The methods of TVA are proffered as more than technical
means for the achievement of administrative objectives. They include
and underline the responsibility of leadership in a democracy to offer
the people alternatives for free choice rather than ready-made prescrip-
tions elaborated in the fastnesses of planning agencies. By 1936, Lilien-
thal had formulated the bases of such a policy.
20 H. A. Morgan, Chairman, TVA, "Some Objectives and End Results of TVA,"
address before annual meeting of the American Society of Agricultural Engineers,
Knoxville, Tenn., June 23, 1941 (mim.).
38 University of California Publications in Culture and Society
This matter of making a choice available, which is the duty of leadership, seems to
me critically important. There are two ways of going about many of these matters.
There, for example, is a steep slope which has been denuded of trees by the farmer.
He has to make a living. He needs this steep slope to grow the things that will keep
his family alive, and so he cuts the trees down and plants his corn, and the soil is
washed off in a few years, and the nation has been robbed of just that much of its
capital assets. . . . Now one way of going about it is to say, "We will pass a law
that any farmer who cuts down the trees and cultivates a slope steeper than a certain
grade is incapable of farming. He is injuring the community and the nation, and by
this law we will take his land away from him and turn it back into forest or meadow."
That is one way. . . . Then there is the other method, which the TVA has pursued,
of giving the farmer a chance to make a choice; recognizing that the farmer does
not cut down those trees because he enjoys cutting down trees or because he likes to
see the soil washed off and destroyed but because he has a problem of feeding his
family and making a living. Give him a choice a free choice by making it possible
for him to use his land in such a way that he will not only be enabled to support his
family but at the same time protect that soil against depredation. This is only one
illustration of many of this conviction I have that a man must be given a free choice,
rather than compelling a choice or having super-men make the choice for him. 21
In this way the Authority has applied a moral sanction to its program
of giving the people and the existing institutions in the area a chance
and the means to participate in an over-all program.
The orientation toward local agencies is also a product of the concep-
tion that the resources of a region include its institutions, in particular
its governmental agencies. The Authority deems it part of its obligation
in connection with resource development that these local governmental
institutions be strengthened rather than weakened, that they be supple-
mented rather than supplanted. In doing so, the Authority directs its
effort toward developing a sense of responsibility on the part of the local
organs and, what is equally important, toward providing them with a
knowledge of the tools available to put that responsibility into action.
The TVA's policy of strengthening local institutions is linked to its
broad responsibilities for regional development. Thus Gordon R. Clapp,
formerly general manager and now chairman of TVA, has emphasized
that maximum regional development is a function not only of the physi-
cal resources of a region but of administrative or managerial resources
as well. 22 This must be evident to all who understand the difficulty of
* David E. Lilienthal, transcript of TVA lecture, Knoxville, Tenn., June 12, 1936
(mim.).
22 Gordon R. Clapp, "The Administrative Resources of a Region: the Example of
the Tennessee Valley," in New Horizons in Public Administration (University, Ala. :
University of Alabama Press, 1945), chap. IV. The publication of this volume, first
of the University of Alabama Press, is itself an indication of the TVA's attempt to
strengthen local institutions. In conjunction with TVA, the universities of Alabama,
Georgia, and Tennessee launched a Southern Regional Training Program in Public
Selznick : T VA and the Grass Roots 39
bridging the gap between a recognition of the needs of a region and the
establishment of suitable methods for their fulfillment. Moreover, the
plans of engineers and scientists depend for their effectuation upon the
decisions and efforts of the people who actually operate upon the soil
and other resources of the area. To attain the cooperation of these people
is a matter not of simple exhortation, but of persuasion and organiza-
tion, of practical attempts through the solutions of their individual prob-
lems to link them with the public goals of the TVA.
A list of agencies with which the TVA has maintained some form
of cooperative relationship includes nearly all of the governmental insti-
tutions in the area : municipal power boards, rural electric cooperatives,
school and library boards ; state departments of health, conservation, and
parks ; state and local planning commissions, agricultural and engineer-
ing experiment stations, state extension services, andTothers. In devel-
oping these relationships TVA has applied the rule that "wherever
possible, the Authority shall work toward achieving its objectives by
utilizing or stimulating the developing of state and local organizations,
agencies and institutions, rather than conducting direct action pro-
grams." 23 In addition, a number of federal agencies, notably technical
bureaus of the U. S. Departments of Agriculture and the Interior, the
Army Engineers, and the Coast Guard, have cooperative arrangements
with the TVA. Notable also are the ad hoc organizations and conferences
which have been established as vehicles for cooperation among the ad-
ministrative agencies within the Valley. These include, among others, a
semiannual conference of directors of extension services and of agricul-
tural experiment stations of the seven Valley states, the U. S. Depart-
ment of Agriculture, and the TVA ; the Tennessee Valley Trades and
Labor Council, bringing together fifteen international unions of the
American Federation of Labor Building and Metal Trades ; an annual
conference of contractors and distributors of TVA power ; and the Ten-
nessee Valley Library Council. Such gatherings help to lay a sound
foundation for regional unity, focusing the efforts of many agencies on
the region as a central problem.
The form of cooperation with state and local agencies varies, but the
pattern of intergovernmental contract has been most fully developed.
Such contracts often include reimbursement by the Authority for per-
sonnel and other facilities used by the state in carrying on the coopera-
Administration which provided internships for ten college graduates in 1944. The
series of lectures published in New Horizons was a feature of the training program.
28 As formulated at an Administrative Conference of TVA staff members, April 21
1943.
40 University of California Publications in Culture and Society
tive program. In many cases, the ideal outcome is viewed as the tapering
off of TVA contributions until, as TVA's responsibilities recede in im-
portance, the local agency carries on by itself. Thus the states have in
some cases begun planning work through their own commissions with
the material help of TVA ; later, state funds have been secured with a
view to continuing the work when TVA's responsibilities for the re-
adjustment of reservoir-affected urban communities would terminate. In
cooperating with the local governments, TVA attempts to establish a
pattern which may be continued after TVA aid has ceased.
The objective of stimulating local responsibility among governments
and associations within the area is basic to the grass-roots approach.
But there are other reasons which support it as sound administrative
policy. The existing facilities of the states, even though they may be
inadequate, are used to capacity, thus avoiding the establishment of
duplicate services and personnel with parallel functions. The TVA is
110 1 ailiSnirtO'n^ve^ts^rywn men in the field and is willing to forego the
prestige that comes from identification as "TVA men" of agents per-
forming services paid for out of TVA funds. The staff is educated to
feel most satisfied when it can show evidence that a local organization
has carried on TVA work and been permanently strengthened by the
experience and in the eyes of its public. In addition, utilization of exist-
ing agencies permits TVA to shape its program in conformity with,.tlje
intimate i knowledffi c^Jocal conditions which such agencies are likely
to have ; at the same time it is possible to restrict the size of the Author-
ity's direct working force.
The attempt to create a working partnership between the TVA and
the people in carrying out a common program for regional development
goes beyond the strengthening of existing governmental agencies, though
this objective is vital. The meaning of the partnership is contained as
well in the use of the voluntary association as a means of inviting the
participation of the people most immediately concerned in the adminis-
tration of the program. In this way, the farmer or the businessman finds
a means of participating in the activities of government supplemental
to his role on election day. If there is fertilizer to be distributed, farmers
are invited, on a county and community basis, to participate in locally
controlled organizations which will make decisions as to the most effec-
tive means of using that fertilizer in the local area. If government land
is to be rented, a local land-use association is organized so that the con-
ditions of rental can be determined with maximum benefit for the com-
munity. If power is to be sold in a rural area, a cooperative provides a
Selznick : T VA and the Grass Roots 41
consumer ownership which retains profits in the community and makes
possible a management guided by community problems and local needs.
If the business area of a city must be modified because of newly flooded
lands, let a locally organized planning commission work out the best
possible adjustment of special interests and long-range planning goals.
Thus, at the end-point of operation, the specific consequences of a federal
program may be shaped and directed by local citizens so that its impact
at the grass roots will be determined in local terms. This procedure is
not only democratic and just, but undoubtedly adds measurably to the
effectiveness of the programs, which will be conjoined to the special
desires of those affected and thus have the benefit of their support
and aid.
The policy of consciously working with and through local institutions
is, in the Authority's view, integrally related to its relatively autonomous
position within the federal system. It is precisely the flexibility accorded
to the TVA management which has enabled it to keep in mind its broad
concept of regional development and at the same time to seize upon
whatever opportunities might arise to implement the concept concretely.
Nationally directed restrictions as to employment of personnel, a host
of regulations framed in national terms, would doubtless greatly restrict
the ability of TVA to establish procedures attuned both to its substan-
tive objectives and to the grass-roots methods by which they are car-
ried out. It would surely inhibit the freedom to search out techniques
uniquely adapted to the special situations of some particular state
government or community if TVA did not have the power to make its
own decisions and to take the initiative in fostering cooperative rela-
tionships. Moreover, the absence of discretion might well be psychologi-
cally decisive in hobbling the TVA staff by binding it to the customs and
traditional modes of action laid down by the broader hierarchy into
which it might be absorbed.
DECENTEALIZATION AND REGIONAL UNITY
While the problem of method is the nub of the TVA approach, the latter
may not be divorced from the program which called it into being. The
outlook of the TVA leadership stresses decentralization as vital not only
for the future of democratic government but also as an indispensable
tool for the effective development of regional resources. The TVA idea
is also a regional idea which looks toward the formulation and execution
of a unified program for resource development; and it is believed that
responsibility and authority for the fulfillment of such a program should
42 University of California Publications in Culture and Society
be allocated to a single agency. This does not mean that a single agency
will actually carry out all the necessary activities, but it does mean that
it will provide over-all direction in terms of regional goals. It means
that governmental initiative, localized and focused in an agency which
has responsibility for unified development, will be available to seize op-
portunities as they may arise, and to make the most of them.
Any general extension of the TVA idea must plainly be rooted in a
recognition of the diversity of needs and potentialities which charac-
terizes the various sections of the United States. This diversity demands
flexibility on the part of federal authority and the use of a method which
can adjust a national program to the special problems of the area of
operation. It is difficult and perhaps impossible to specify in any exact
way how the lines may be drawn around an area to say this is a Region. 2 *
That will have to be done in terms of the existence of a problem area,
or a resource base, or some other criterion or combination of criteria
which will serve to reflect the practical unity which ordinary men under-
stand when they associate together for regional objectives. What is
important is not precise specification or neat boundaries, but the recog-
nition of focuses of potentiality and need which, treated as wholes, can
contribute most effectively to regional development and thereby to na-
tional welfare. Diversity in nature and tradition requires diversity in
program and policy ; for the idea of regional administration is frankly
opposed to those forces which strive to introduce uniformity into all
aspects of government programs.
The decentralized regional agency offers a means of creating a center
of regional responsibility, planning, and coordination within the frame-
work of the existing federal system. Through it, functions too broad to
be undertaken by a single state and yet not actually national in scope
may be initiated at the proper level and given direction and scope in
terms of the special problems of an area, at the same time that national
policy may be brought to bear upon regional needs. Such staff agencies
as the Bureau of the Budget and a National Resources Planning Board
can, as integral parts of the Presidency, direct from the national govern-
ment matters of broad perspective. The decentralized regional agency is,
however, not the same as the regionalization of national agencies through
the establishment of field offices. Such federal outposts are not charged
with responsibility for integral regional development ; and, equally im-
portant, regional unity demands the continuous development of a pro-
2t Cf. D. E. Lilienthal, TVA : Democracy on the March (New York: Pocket Books,
1945), p. 167.
Selznick: TVA and the Grass Roots 43
gram based on study and decision in the field. A truly regional agency
is multifunctional and single purposed. It has the means and the author-
ity to engage in or to initiate several major and many lesser projects in
accordance with the developmental needs of the area. At the same time,
the outlook of the agency is integral, devoted to the unified and conserva-
tive exploitation of the region as a whole.
In the TVA, the idea of unified resource development is associated
with a characteristic spiritual lesson. Largely under the influence of Dr.
H. A. Morgan, former President of the University of Tennessee and one
of the original members of the TVA Board, the agency's outlook has
been framed in terms of the conception of a "Common Mooring," linking
industrial and governmental activities to a fundamental human respon-
sibility for the prudent utilization of natural resources. 25 "Dr. H. A.'s"
lectures are well known in the Authority, and they are presumably a
useful method of communicating the unified outlook to the staff. This
Common Mooring consists of the basic factors provided by nature which
make organized life possible : the inexhaustible elements in the air,
water, and soil and the exhaustible resources which a morally guided
intelligence must husband :
It has not been an easy task . . . for nature throughout the millions of centuries to
keep the elements and compounds within the evolutionary program so that when man
should appear he would find a universe of complex interacting interdependent ele-
ments and compounds at his service. These have been so governed by laws that man's
environment has permitted not only his existence, but that of all plants and other
animals. This complex has definitely challenged man to discover his relation to his
Creator and the creative process through his relation to his environment and to the
mineral, plant, and animal kingdoms in which he found himself. This challenge has
not stopped here. Over the centuries there has developed within man a sense of respon-
sibility and moral obligation to assist nature in augmenting, through intellectual
concepts not available to any other beings, the program that brought him into being. 20
The TVA program is viewed as an application of the Common Mooring
to regional development. It is a "single-purpose approach to many-sided
problems," treating as fundamentally interrelated the basic, inexhausti-
ble resources : air, water, and land. The unified approach made possible
by a single agency is a needed antidote for the division of the universe
made necessary by specialization. Put another way, TVA makes for
25 See Ellis F. Hartford, Our Common Mooring, prepared for the Advisory Panel
on Eegional Materials of Instruction for the Tennessee Valley (Athens, Ga. : Univer-
sity of Georgia Press, 1941).
28 H. A. Morgan, "The Common Mooring," in Applications of the Common Moor-
ing, compiled and edited by Howard P. Emerson, Chairman, subcommittee on Presen-
tation of the Common Mooring, Advisory Panel on Regional Materials of Instruction
for the Tennessee Valley, Knoxville, Tenn., June, 1942 (hectographed), p. ii.
44 University of California Publications in Culture and Society
catholicity of outlook-.jorjthe leadership of the single agency_acts as jin
integrating force in educating the comppnen^parts of its staff to view
the problem of resource development as a whole.
The existence and the program of a regional agency is tied to the
definition of a resource base. This will not be the same for all areas, but
presumably it is possible to specify some key resource whose develop-
ment can be the hub of a general program for each area. The decen-
tralized agency should be made directly responsible for the development
of the key resource, and be given the authority to make such arrange-
ments with other governmental agencies as will most effectively fulfill
the over-all requirements of the region. With these powerful tools, the
regional agency can become an effective coordinator, mobilizing all pos-
sible participants in the interests of an integral plan.
A regional agency is the reflection of the physical unity of the resource
base of the region. In the Tennessee Valley, the control of water in the
river channel is bound to the control of water on the land. The river
requires a unified system of dams and reservoirs ; the land requires the
development of sound agricultural practices and the wider use of phos-
phatic fertilizers. These jobs require different techniques and specialized
personnel, but the development of the area demands that they be seen as
a whole, as they affect each other. A regional agency can undertake to
bring those perspectives together ; it can also take into account the social
consequences of its programs as the TVA does when it aids communities
to intelligently readjust themselves to newly created reservoirs, and
flooded-out farmers to relocate. The interrelation of social phenomena
makes for a spreading network of consequences following in the wake of
action ; as a result, a special public emerges. It is this public, plus the
national interest ultimately involved, which forms the political basis for
the decentralized structure of the regional agency.
So runs the official doctrine.
It will be noted from the above account that the concept of administra-
tive decentralization or grass-roots administration, as used by TVA,
refers to three administrative levels : *
1. The jocation of administrative control in the area of operation,
with^the^ Authority as a wEole, in relation 10 thA fMdf&l IJOVCrflffieTit,
taken as an example.
2. The carrying on of operations with and through existing institu-
tions already organized in the area of operation. The relation of the
Selznick : T VA and the Grass Roots 45
Authority to the agricultural extension services of the land-grant col-
leges is one of the imp^tanTexampTes oTthTs procedure^
3. The participation of local people at the end point of administration
of the program, for example, through county soil improvement associa- t
tions set up in connection with the TVA fertilizer test-demonstration
program.
In the following pages an attempt is made to subject the official doc-
trine to functional analysis, and to set forth in detail the operation of
one important phase of the grass-roots procedure, tracing its conse-
quences for the role and character of the TVA. The major direct concern
of the study is with level 2 noted above, secondarily with level 3. How-
ever, it will be evident in the course of the analysis to follow that the
policy pursued by TVA in its own area cannot be free of consequences
for its relation to the federal government, and therefore for level 1.
CHAPTER II
/
THE FUNCTIONS AND DILEMMAS OF
OFFICIAL DOCTRINE
The grass-roots policy is merely a rationalization. It is absurd to have a federal agency
trying to work through organizations which place it at a second remove from the
people> A TVA STAFF MEMBER (1943) : AN EXTREME VIEW
THE PARAPHRASE of official doctrine, briefly stated in the preceding
chapter is not, of course, complete. Ideas sincerely held and highly
prized are not likely to be acceptably presented by a mind alien or even
detached. The leadership of the TVA would be quick to point to sub-
tleties unnoticed, ramifications ignored, and meanings crudely put. Such
an attitude, if expressed, would be readily understandable. For this is a
doctrine to which the TVA leadership is deeply committed. The idea
of a grass-roots administration is, as we have already noted, no casual
or minor element in the consciousness of the Authority's staff. It is, on
the contrary, one of the symbols most frequently referred to inside the
organization. Speeches to new employees, letters to information seekers,
a popular book written by the chairman, memoranda discussing projects
and procedures, all attest to the importance of the grass-roots idea in
TVA. The systematic promulgation of these ideas helps to define the
character of TVA as an organization and serves to shape the outlook
of its staff. Open though it is to various interpretations, there can hardly
be any question that, at least in the regional development departments,
the grass-roots approach is accepted without question as official policy
and for the most part is warmly endorsed as effective administrative
procedure.
SOURCES AND FUNCTIONS
Among the many and pressing responsibilities of leadership, there arises
the need to develop a Weltanschauung, a general view of the organiza-
tion^ position and role among its contemporaries. For organizations are
not unlike personalities : the search for stability and meaning, for secur-
ity, is unremitting. It is a search which seems to find a natural conclusion
in the achievement of a set of morally sustaining ideas, ideas which lend
support to decisions which must rest on compromise and restraint.
Organizations, like men, are at crucial times involved in an attempt to
close the gap between what they wish to do and what they can do. It is
natural that, in due course, the struggle should be resolved in favor
[47]
48 University of California Publications in Culture and Society
of a reconciliation between the desire and the ability. This new equili-
brium may find its formulation and its sustenance in ideas which reflect
a softened view of the world. The ethic of revolt, of thoroughgoing
change, assumes that human and institutional materials are readily mal-
leable and that disaffection from current modes of thought and patterns
of behavior can be long sustained in action. But leadership must heed
the danger of strain and disaster as recalcitrance and inertia exert their
unceasing pressures; in doing so, it may see wisdom in the avowal of
loyalty to prevailing codes and established structures. The choice, indeed,
may often lie between adjustment and organizational suicide. 1
This quest for an ideology, for doctrinal nourishment, while general,
is uneven. Organizations established in a stable context, with assured
futures, do not feel the same urgencies as those born of turmoil and set
down to fend for themselves in undefined ways among institutions
fearful and resistant. As in individuals, insecurity summons ideological
reinforcements. The TVA was particularly susceptible to this kind of
insecurity because it was not the spontaneous product of the institutions
in its area of operation. Thus one Southern critic, Donald Davidson,
wrote :
The regional psychology of the TVA conception is not native to the South. The TVA
was not created in response to a Southern crusade although some Southern Con-
gressmen had long agitated for a final disposition of Muscle Shoals. There was no
popular outcry in the South for a TVA. The sponsor of the TVA Act was a Middle
Westerner, Senator Norris, who was crusading against the electric utilities. . . . The
five states concerned had no opportunity to debate the project or to contribute ideas
or leading personnel, or by any direct means to make known their opinion, if they
had any; the project was superimposed (however benevolently) upon them. 2
The often-remarked change in attitude toward TVA among people of
the Valley from hostility or suspicion to acceptance and support
lends weight to Mr. Davidson's observation. In 1946 a reporter for the
Denver Post wrote from Knoxville :
The TVA is such a revolutionary type of government organization that its entrance
into this generally conservative, middle-south region thirteen years ago was marked
by many bitter controversies, the wounds of which, however, have mostly healed now.
1 For those to whom the given organization is a vital instrument for personal satis-
faction, the outcome is virtually inevitable. One division chief pointed out that, in the
TVA, the group of younger men whose careers were bound up with the fate of the
Authority was far more ready to adapt itself to new policies framed in terms of
"realistic" adjustment than was the earlier leadership, the reputations of whose
members were already established.
2 Donald Davidson, "Political Eegionalism and Administrative Regionalism,"
Annals of American Academy of Political and Social Science, 207 (January, 1940)
141.
Selznick : T VA and the Grass Roots 49
From small businessmen, school teachers, farmers, newspaper writers, city officials,
railroad workers, and professional men, I have gained the impression that those who
still oppose TVA here are in the minority, decidedly. ... A Knoxville businessman :
"This is a conservative community. This is the heart of the Bible Belt. Our people
voted down Sunday movies, for example. When TVA set up headquarters here with
flocks of 'bright young men' many of whom were not yet dry behind the ears, our
people resented it and referred to them as 'those foreigners.' Now the resentment is
gone. The socially minded theorists in TVA who couldn't deliver in actual practice
have gone." 3
The TVA could not take its position for granted. It had to feel its way.
This required the formulation of a policy that would reassure external
elements and would so educate its own ranks as to maximize the possi-
bilities of social acceptance.
But adjustment is a complex process. In pursuing an analysis of this
sort it is essential to remember how diverse and unstable are the sources
and functions of any set of guiding ideas with which an organization be-
comes identified. Some organizations are born of ideological conflict, and
administered by the victors with a view to furthering "the cause." If so,
the doctrine espoused by the leadership may be understood only with
reference to the symbolic status of the organization. Such might have
been true in TVA if planning as such or positive government or some
other component of the doctrine of the liberal-socialist movement in the
United States had been emphasized by the directors as the special TVA
philosophy. Then the problem of theoretical interest might have been to
examine the possibilities of a gap between what men said (in the light
of doctrinal commitments) and what men did (under the pressure of
practical exigency, the force of circumstance). But there are other
sources of ideas, sources close to the special problems of organization as
such, which require some attention.
One such source of official doctrine derives from the simple fact that
formal organizations live in institutional environments. Each is a co-
operative system, and involves its members wholly, with all the cultural
predispositions and tangential commitments given them by lives apart
from the formal purpose of the organization. Thus the sheer weight of
special education provided by the community from which personnel is
drawn will shape and limit the ideas which can be developed by the
organization. Moreover, there are specific local pressures, especially of
the various client groups served, which infuse the organization with new
goals and interest-defensive ideas.
3 L. A. Chapin in Denver Post, May 3, 1946. This article was part of a series
entitled "Denver Post Survey of TVA."
50 University of California Publications in Culture and Society
In addition, ideas arise from the need for internal communication as
a means of developing organizational unity and homogeneity. This func-
tions both to present a consistent interpretation to the world, especially
when subject to attack, and to establish a foundation for the smooth flow
of directives, the ready acceptance of executive decision. This unity,
when based on the cultivation of an appropriate set of attitudes, provides
a framework for the development of special policies which will be at-
tuned to the basic view of the leadership. The organization takes on the
aspects of a social organism, with a set of precepts that will be taken for
granted. This has useful administrative consequences :
1. If a basic point of view is laid down and integrated into the psychol-
ogy of the second and third level leadership, if not farther down the
ranks, the possibilities of decentralization without injury to policy are
vastly enhanced. It is almost axiomatic that in new organizations, in
which the adherence of the administrative personnel to the viewpoint
of the directorate is not dependable, measures of formal control from
the top must be introduced. When, however, indoctrination of official
policy has been sufficiently extended, formal controls may be loosened.
Ideas and the attitudes they foster may serve as surrogates for a system
of rules and formal discipline. Thus in the early days of the TV A, before
the adoption of the Employee Relationship Policy in 1935, a Labor Rela-
tions Division handled grievances directly with employees. An employee
could by-pass the supervisory hierarchy in his own division and contact
"Labor Relations" which had the power to investigate and force settle-
ment. This control through an organ of central management was con-
sidered necessary because in the beginning, according to a staff member,
TVA had "many supervisors, high and low, fresh from the notoriously
antilabor construction industry. Hence the wielding of a big stick was
thought necessary."
2. The educational function common to all such structures of pain-
lessly but effectively shaping the viewpoints of new members is facili-
tated, thus informally but effectively establishing the rubrics of thought
and decision. This is well understood in practice, reflected in the use of
organizational labels (a "Forest Service" man, an "agriculturist," etc.)
so that special attitudes and characteristic administrative methods may
be identified.
3. A consistently developed doctrine providing a sense of unity to the
organizational endeavor is particularly useful in binding together tech-
nical experts who otherwise tend to emphasize their special professional
interests. Formal methods of coordination may be buttressed by the com-
munication of a unified outlook, for the latter will tend to make the
Selznick: TVA and the Grass Roots 51
technicians think in terms of the organization as a whole. In the TVA,
the problem of coordinating the views of experts has been considered
especially important by central management. The ability of a repre-
sentative of one division to appear before the Board of Directors repre-
senting not only his own but some other division is hailed as a significant
achievement. Such a welding of viewpoints is aided by the coordinating
power of a well-developed set of ideas.
4. A further source of organizational doctrine is the normal tendency
for an agency to defend its own functions. To some extent, even the
leader of a subordinate division is expected to speak for and defend the
interests of his division. As a matter of fact, his role is denned by
the system as one of representation, not without reason. It is convenient
and perhaps necessary for a central coordinating agency to circularize
its staff components for opinion as to policy; the significance of such
opinions lies in setting forth the special needs and .outlooks of the con-
stituent divisions. Coordination requires such a collation as the raw
material for decision that will take all factors into account. Moreover,
the possibilities of competition among subordinate sections are thereby
furthered ; this has an incentive value and to some extent is an indica-
tion of the vigor with which a program is being pushed, although degen-
eration into bureaucratic politics may and does occur. The subleader
must also see that the needs of his staff are represented when decisions
affecting it are made, 4 a function which will turn his attention inward
to emphasize his organization as a value in itself. Professionally, too,
the personnel becomes identified with the agency, deriving prestige and
disesteem from the fortunes of the organization itself. To defend the
organization is often to defend oneself. These defensive activities are
aided when a set of beliefs is so fashioned as at once to fortify the special
needs or interests of the organization and to provide an aura of disin-
terestedness under which formal discussion may be pursued. 5
4 See Donald C. Stone, "Notes on the Governmental Executive," New Horizons in
Public Administration (University, Ala. : University of Alabama Press, 1945), p. 58.
5 Of course, the elaboration of a doctrine is not the only avenue to, or warrant of,
solidarity in formal organizations. Personal loyalties, habit and custom, as well as
the tendency of organizations to function as alter egos, are among the many sources
of informal solidarity. It should also be pointed out that this organic quality has other
consequences not always considered useful, among them a rigidity which makes it
difficult to change or elaborate the formal purposes of the organization. Hence new
functions, though in the same program area, may require new organizations. Never-
theless, no administrative leadership may overlook the values for internal communi-
cation of developing a unified outlook on the part of key personnel. See C. I. Barnard,
The Functions of the Executive (Cambridge: Harvard University Press, 1938), p.
87 : "The inculcation of belief in the real existence of a common purpose is an essential
executive function. It explains much educational and so-called morale work in po-
litical, industrial, and religious organizations that is so often otherwise inexplicable."
52 University of California Publications in Culture and Society
The considerations just noted refer to the role of any official doctrine
in an organization. However, in pressing our inquiry into the functions
of the T VA's grass-roots doctrine, it is necessary to examine its origins, in
particular whether the doctrine arose from within the Authority or
was provided for it at the outset. This is important for an analysis which
attempts to explain the function of the doctrine by the special needs of
the organization as an adaptive structure. 9
The elaboration of the grass-roots policy as official organizational doc-
trine seems to have filled what amounted to an ideological vacuum. The
formulation of this doctrine did not precede the establishment of TVA,
nor did its precepts materially influence the nature of the organization
created by the TVA Act. The form of the TVA organization was related
more to the specific political situation at the time of its inception than
to any broad principles of public management. To be sure, the advan-
tages of a corporate organization having the powers of government and
the flexibility of private enterprise a form already familiar in Ameri-
can government were apparent; but the key elements of the situation
in 1933 which seemed to point to the need for an autonomous agency
were (1) the struggle over public power, and (2) the experimental
character of the TVA as a planning agency. Neither implied a self-
conscious doctrine such as that developed by the Authority.
1. An autonomous agency, having broad powers of discretion, was an
appropriate device for organizing a battle against the electric utility
interests. A leadership could be chosen which understood its responsi-
bilities in terms of the national struggle undertaken by the Roosevelt
administration. The autonomous directorate could muster its forces
freely, according to the needs of the conflict, without being bound by
rules formulated for extraneous ends and without being hampered by
the customs and prior commitments of a federal department. Thus
we have already noted (see p. 31 above) that the demand to be free of
U. S. Civil Service Commission control was supported by Senator Norris
because of his fear that "spies" of the "power trust" might infiltrate into
6 The analysis of origins, while relevant here, is not crucial. Any ideas maintained
and elaborated by the leadership whatever their origins may fruitfully be sub-
jected to functional analysis. However, the significance of such an analysis is
strengthened if the origin as well as the meaning of the doctrine is referrable to
the needs of the organization. This does not mean that the motives of its espousers
are called into question : we can tell very little about them, and it is fair to assume
that Lilienthal and those around him were activated quite sincerely by a deep con-
cern for democracy as a moral problem. Let it be noted that our entire analysis is
rooted in an interest in the meaning of events, stressing such terms as "function"
and "consequence." This is quite aside from the consciousness or moral quality of the
participants.
Selznick: TV A and the Grass Roots 53
the organization and cause "great injury and sabotage." Again, flexi-
bility allowed the TVA in the use of its funds was justified partly on
the basis of the electric power controversy. This flexibility would make
it possible for the agency to seize opportunities as they arose for advanc-
ing the cause of public power. For example, a local electric cooperative
whose area was threatened with invasion by private utilities might be
given aid by the Authority without undue hindrance or delay.
2. The TVA Act, and more fully the message of President Roosevelt
requesting the legislation, represented a political challenge, even if it
was not an entirely new departure in the exercise of federal authority
for general welfare objectives. According to the President, the TVA
was to be "charged with the broadest duty of planning for the proper
use, conservation, and development of the natural resources of the Ten-
nessee River drainage basin and its adjoining territory for the general
social and economic welfare of the Nation." 7 In the public mind, the
meaning of TVA was bound up with the values of planning and social
responsibility as corollaries of the public-power controversy. This repre-
sented a model of positive government which could not be put forward
for the federal government as a whole but rather, for reasons of strategy,
as an experiment to be extended as conditions might warrant. In that
capacity, and separated from the existing federal apparatus, the agency
would remain in the national spotlight as a living example of what
planning could accomplish. Moreover, the Authority still remains the
receptacle of the hopes of those who stand for national planning; its
autonomy is defended in political terms, on the basis of its program
rather than the grass-roots method.
It seems clear that in 1933 the implications of the managerial form
of the TVA for the future of federal administration were not well under-
stood. Certainly the problems of public power and planning were press-
ing, and no systematic doctrine, such as the grass-roots theory later put
forward by the Authority, was envisaged. It is true that the Authority
was authorized to cooperate with local and state institutions, but no
positive program of strengthening existing institutions was outlined
as part of the conception of the TVA. The Authority was also authorized
to cooperate with federal agencies, but this has not been stressed by
TVA as a basic objective ; Lilienthal has suggested that regional agencies
should be compelled to cooperate with local and state agencies but has
not recommended a similar injunction with respect to organs of the
federal departments. In general, TVA is willing to work with federal
7 House Doc. 15, 73d Cong., 1st sess. (1933).
54 University of California Publications in Culture and Society
agencies, but emphasizes working through state and local agencies. In
relationships with federal agencies, the grass-roots policy is pressed:
while there is little difficulty in getting along with federal agencies which
work through established state and local channels, where they do not,
TVA considers that it has a responsibility to "make a point of method,
trying to persuade the federal agency to modify its method." 8 This
special point of view was developed by the Authority itself, and in its
systematic form represents a unique approach to the administration
of federal functions. There is some indication, moreover, that the TVA
theory was not accepted as an administrative pattern by President
Roosevelt. In his message of June 3, 1937, in support of the Conservation
Authorities Bill, the President specified that "projects authorized to be
undertaken by the Congress could then be carried out in whole or in part
by those departments of the Government best equipped for the purpose,
or if desirable in any particular case by one of the regional bodies."
This clearly did not accept the grass-roots approach of the TVA, which
calls for the decentralized execution of federal programs as a matter of
fundamental policy.
If, as has been suggested, the special administrative form of TVA was
derived from the circumstances of its inception rather than from general
principles, it should be easy to see that an ideological vacuum would
arise soon after the initial period. After victory was assured, and it
became clear that the TVA was to endure, it was necessary for the leader-
ship to develop some notion of its place within the American political
system. Somehow an answer had to be found to the question : What now ?
Is the Authority a viable method of administering federal functions on
a permanent basis, apart from the context of conflict and experiment ?
It was natural though perhaps not inevitable that the leadership
should look with pride at the organization which had won so much praise,
and natural that it should feel that the TVA method had an enduring
value as effective management in a democracy. To some extent, fondness
for the grass-roots theory resulted because it was raised as a banner in
the internal conflict which rocked the agency not long after its incep-
tion. 10 A unified outlook was not known until the removal of Arthur E.
8 As formulated at an administrative conference, Knoxyille, July 14, 1943.
9 See Hearings on S. 2555 (Conservation Authorities Bill), Senate Committee on
Agriculture and Forestry, 75th Cong., 1st sess. (1937), p. 2.
10 The history of this struggle is detailed at length in the Hearings of the Joint
Congressional Committee to Investigate the Tennessee Valley Authority (1938). The
meaning of the split, in terms of its consequences for policy within the organization,
has never been fully recorded. It is especially unfortunate that a study made by Pro-
fessor Herman Finer of the administrative history of the Authority, undertaken in
1937 under the auspices of the Social Science Eesearch Council, has remained un-
available.
Selznick: TVA and the Grass Roots 55
Morgan by President Roosevelt. The struggle which led to that removal
is often interpreted as having involved the question of whether the TVA
would work with the people of the Valley or impose its program despite
them. However that may be, it is clear that originally there was no unity,
and that the special outlook of Harcourt A. Morgan and David E. Lilien-
thal was formulated in the course of controversy. The latter is always an
effective environment for the crystallization of loyalties to a certain set
of ideas.
The basic functions of the official TVA doctrine have been (1) inter-
nal, satisfying such needs as effective communication, and (2) adjust-
ment to the area of operation. The content of the doctrine is not, perhaps,
of great significance for the former, but it is decisive for the latter. What-
ever the specific formulation, the idea of taking account of, or working
with, or otherwise accepting the existing social institutions of the Valley
is clearly consonant with the iron necessities of continuing to exist and
work in that area. Whether or not the TVA had developed the grass-
roots theory, it would have found it necessary in some way to get along
with those forces which might wreck the program if moved to resistance.
In relation to the states, counties, and other local agencies, the TVA
could not have operated successfully without framing its program within
the existing pattern of government, including the powers and the tradi-
tional prerogatives of the local units. That must be taken for granted
by any organization, public or private, which seeks to accomplish a
large-scale program within a populated area. The alternative is force
an alternative which even occupying armies in enemy lands are loathe
to use. In the field of economics, too, the TVA Commerce Department
must assume that the main industrial base in the Valley will be private
industry. Its work must be based on that fact, so that it may convince
existing industry of the value of good stewardship of the Valley's re-
sources. Thus in one sense the grass-roots approach simply verbalizes
an administrative approach which any agency would follow of necessity.
The formulation of a doctrine placing a halo over procedures which
might in any case be considered merely normal and necessary, functions
to clear away the suspicions and latent resistance with which an agency
imposed from above might be greeted. Such a doctrine also provides a
vehicle for the justification of special efforts which may be made to
placate the local leaderships and give them a stake in the fortunes of an
organization not their own. Thus frontal appeals by local government
to the TVA for favors may be avoided if the Authority itself systemati-
56 University of California Publications in Culture and Society
cally exploits opportunities to assist these units with advice and money.
The theory serves also as a means of organizing a "mass base" for the
Authority by formulating and justifying aid to labor unions, coopera-
tives, farmers, farm organizations, universities, and state governments.
Such a base serves as a protection if the agency is subject to attack.
Though emphasizing again that we do not discount the sincerity of the
TVA leadership, it is clear from the record that the Authority expects
support from those agencies with which it works. Thus when, in 1941,
the President of the University of Tennessee publicly criticized the pro-
posed construction of Douglas Dam, and seemed to call into question the
general benefit of TVA for the Valley, the Authority was much dis-
pleased, and threatened to "reconsider" the method by which it carried
on its agricultural program through the land-grant colleges. When it is
understood that the method in question is offered as one of the prime
examples of the grass-roots approach, the significance of this can easily
be appreciated.
The official doctrine does not appear to fill all of the TVA's needs for
adjustment. Among the most important of the gaps recognized by mem-
bers of the staff is the need to adjust the organization to the existing
structure of the federal government. That, too, is part of the Authority's
environment, though during the tenure of President Koosevelt the need
to adjust the TVA to the federal structure may not have been as impor-
tant for the survival of the organization as adjustment to local institu-
tions. At least in terms of the development of TVA doctrine, the need
of adjustment to the area of operation has been taken as basic. For the
TVA has failed to develop a formulation which would be acceptable to
the administrators in Washington, and in a practical sense there are
areas of maladjustment which remain unreconciled. One TVA staff mem-
ber has said that "when TVA was getting started it was perhaps neces-
sary to avoid getting sucked into the federal stream, but now the
Authority must develop effective working relationships with the old-line
agencies. . . . TVA's relations to federal agencies have produced some
enemies in the administrative setup, throwing anti-TVA barbs in the
fight in Congress." Kelationships with some long-established agencies in
Washington are very good, but it is clear that the TVA's role in the
federal structure has not been satisfactorily formulated. A Washington
"bureaucrat" has written :
. . . most of the discussions concerning possible forms of organization for resource
development have largely ignored the possible effects of projected forms of organi-
zation on the general governmental framework as well as relationships with the other
Selznick: TV A and the Grass Roots 57
agencies in this framework. Mr. Lilienthal's article (N. Y. Times Magazine, Janu-
ary 7, 1945) is devoid of any reference to this problem. 11
It appears, indeed, that TVA has not faced the need to accept the federal
structure as given, and has taken on an "all or none" attitude whose
revolutionary implications for the structure of the national government
are accepted, though seldom made explicit. This viewpoint is irrespon-
sible in the nonderogatory sense of a single-minded pursuit of special
goals which leaves consequences to be handled by those directly affected
and by the compromises of history.
There are few organizations which can afford the luxury of a neat or
complete correlation of ideas and action. The needs which lead to or rein-
force the formulation of an official doctrine do not ordinarily exhaust
the demands made upon an organization. To be sure, there must be some
warrant in action for the general ideas propounded by the leadership,
but the latter is free to select from a portion of its program a set of ob-
jectives which will reflect its primary aspirations. This fragment may
become the receptacle of meaning and significance and on the level of
doctrine, of verbalization may infuse the organization as a whole with
a special outlook. But only when a leadership is reckless of the fate of
its charge may it drive toward the exemplification of an administrative
ideology in all phases of its activity. The TVA has been no exception. Its
needs are not simple and are certainly not exhausted by those which have
generated the grass-roots theory. For example :
1. In the light of its commitments to press the struggle for public
power, the TVA could not permit its power program to be channeled
through any of the existing agencies of the state governments in its area.
Although TVA has acted primarily as a power wholesaler, it is fair to
say realistically that the power program has been one of "direct action."
It is true that the power distributors are local municipalities and elec-
tric power cooperatives, and these are counted as grass-roots agencies,
but the difference between the operation of the electric power program
and the agricultural program is too great to be designated by the same
name. In power, the TVA carries forward its own program, with strin-
gent contractual controls. In agriculture, the TVA has shaped its pro-
gram to use the existing agricultural leadership. The power cooperatives
have been built (though based on local response) with the direct and
forceful intervention of the TVA itself. Such means have been ruled
out in the agricultural field.
u William Pincus, "Shall We Have More TVA's?," Public Admin. Bev., V (Spring,
1945). Cf. also C. Herman Pritchett, The Tennessee Valley Authority (Chapel Hill:
University of North Carolina Press, 1943), pp. 219 ff.
58 University of California Publications in Culture and Society
2. In other fields the TVA has not always waited for the development
of proper institutional channels on the local level. In the building of
the river terminals, an important phase of the Authority's responsibili-
ties in the development of navigation, local agencies were not always
considered adequate. The decision to build a river terminal is made by
the TVA Board, and it is difficult to determine the criteria by which the
local channels are rated as adequate. Sometimes, notably in local plan-
ning, new agencies have been organized, with TVA occasionally instru-
mental in the initiation of new state legislation.
3. In 1942, during the wartime emergency, the TVA went ahead with
the construction of Douglas Dam despite considerable protest from the
area. Whatever the merits of the case for building the dam, it is evident
that in this important matter opportunities for local decision were not
available. While the decision to build the dam was made in conjunction
with the leaders of the national government, there is no evidence that
the TVA Board opposed the development. In this matter, as in many
others, the TVA represents national policy and executes that policy with-
out regard to local interests. Despite the consequent difficulties for ad-
justment to the area of operation, it would appear to have been consonant
with the broad interests of TVA to produce the maximum possible power
during wartime to bolster its position on a national scale. 12
Beyond the fact that an official doctrine, however sincerely held, is
not usually operative in all phases of an organization's activity, it is
worthy of note that ideas may serve multiple needs. The multiplication
of the functions it serves may strongly bolster the doctrine because of
the ease with which it comes to hand as repetitive problems are faced.
For it is little more than a truism that the utility of an idea not so much
in inquiry as in smoothing the path of a predetermined course of action
^ may easily invest it with a specious cogency, drawing attention away
;from objective tests and evidence. Thus the theory of "unified resource
development" may (1) aid in breaking down the barriers which sepa-
rate technicians within the regional authority, and (2) serve as a justi-
fication for control of resource development by a single agency. Again,
"water control on the land" is an effective formula which ties general
welfare objectives such as the improvement of agriculture and soil con-
servation to the constitutional pegs of navigation and flood control ; but
the concomitant idea of carrying on soil conservation in order to protect
12 Like all such statements, this seems to imply a Machiavellian motive. That is not
intended ; nor is it suggested that this is the sole, or indeed, any "reason" for the TVA
action. The event has, however, an objective meaning and significance for the TVA's
organizational strength.
Selznick : T VA and the Grass Roots 59
the investment of the government in dams and reservoirs also serves to
justify exercise of those functions by TVA rather than some other
agency. A formula, if it is sufficiently pat, may be a substitute for in-
quiry, a possibility which is strongly reinforced if the formula also pro-
motes a resilient self -consciousness.
UNANALYZED ABSTRACTIONS
Language used for self -protection and for exhortation develops terms
which are unanalyzed, and persistently so, for their effectiveness de-
pends upon the diversity of meanings with which they may be invested.
This is well understood in the field of frankly moral injunction, emo-
tionally formulated, designed to persuade. But the similarity between
the glittering generalities of the sermon or political harangue and the
apparently technical devices of administrative relationship, while sig-
nificant, is far less obvious.
This similarity lies in the fact that all statements contain unanalyzed
terms until and unless a specifiable context is at hand. Such ideas as
"to achieve unity" or "lend assistance" assume meaning empirically
only when we know how to answer the questions : with or to whom ?
under what circumstances? and for what? There are times when cir-
cumstances will permit only one answer, but normally a process of
selection is involved which, when made explicit, provides the abstrac-
tions with a concrete reference. Usually, when methods are essentially
routine, the implicit selection is understood by the participants, so that
what seem to be unanalyzed abstractions have contexts adequately
specified in precedent. But when formulations of procedure are elabo-
rated into organizational doctrine, and come to have a value and function
apart from technical goals, the doctrine becomes a vehicle of organiza-
tional self -consciousness, a tool for the education of the ranks, and may
become progressively divorced from any definite subject-matter. In
the process of being informed with an affective content, the procedural
doctrine may be linked to some commonly accepted symbol such as, in
America, "democracy" in order to share a derived and conditioned
aura. Increasingly, concrete analysis will be avoided, and since un-
analyzed abstractions cannot guide action, actual behavior will be
determined not so much by professed ideas as by immediate exigencies
and specific pressures.
It has been suggested that the Tennessee Valley Authority grew to
organizational maturity under a natural pressure to come forward with
a theory which would justify the very nature as well as the existence
60 University of California Publications in Culture and Society
of the organization. For TVA was from its inception a challenge not
only to the "power trust," but among other things to the theory of
government which prevailed in its area of operation : the idea of local
sovereignty and "states' rights." Objectively, perhaps not always ex-
plicity understood, it became the task of the Authority's leadership to
find a means of mitigating the opposition to an organization imposed
upon the area by the central government. This task was a social need,
essential to the continued existence of the agency. Such a need does not,
of course, mechanically determine the flow of ideas ; but it creates prob-
lems and possible consequences for the life of the organization which
select and reinforce convenient formulations, and which subtly reject
those which carry an implicit challenge or which may tend to undermine
essential loyalties.
It need not be denied that the notion of a grass-roots administration
is a product of the TVA's genuine concern for democratic procedure ;
that concern may be associated with the fact that the specific doctrine
produced has been shaped by the need to adapt the organization to its
institutional environment, to come to terms, to make its peace with the
existing social structure. These objectives can, in a limited sense, go
hand in hand, the one fortifying the other. But the limits are soon
reached. For the needs of maintaining the organization tend to drive
it toward alliances and mechanisms of participation which are specific
and immediately most effective. Commitments are made, traditions and
habits laid down, which have inevitable restrictive pressure on the
exercise of democracy. As a consequence, there is a strong tendency for
the theory itself to contain unanalyzed elements, permitting covert
adaptation in terms of practical necessities.
The most obvious of the unanalyzed elements in the grass-roots theory
of administration is the use of such terms as "the people" and "institu-
tions close to the people." Returning again to a formulation by Lilienthal,
we may read :
To administer national laws entirely from Washington . . . tends to exclude and
ignore the local and State institutions the people have already set up. Usually these
agencies are close to the people and understand their problems; they can often be
of great value in helping to carry forward national policies. But in the nature of
things an overcentralized, remote administration of Federal functions will be unable
or unwilling to enlist the active partnership of these existing institutions, public
and private agencies such as the universities and extension services, local and State
planning commissions, State conservation boards, chambers of commerce, boards
of health the list is long and inclusive. 18
13 D. E. Lilienthal, "The TVA : A Step Toward Decentralization," an address before
the University of California, Berkeley, Calif., Nov. 29, 1940 (mini.).
Selznick: TVA and the Grass Roots 61
LilienthaPs theory attempts to adjust the new requirements of public
policy to the traditional federal system. Without desiring or trying to
reverse the trend toward increased federal authority, he aims to so ex-
ercise that authority that the existing state administrative structure
may be fitted into the new pattern. In the above quotation we find
specified not only the need for decentralized administration, but its
avenues as well. The administration of federal authority is to use "exist-
ing" agencies "close to the people." The assumption is made that these
institutions are in fact representative of the people, or that in the long
run they will most closely approximate representativeness. In those
circles in the Authority which defend the official viewpoint, the tend-
ency is to speak in terms of the "long run" and to make little effort to
defend the existing character of the local organizations on the basis
of democratic criteria. In fact the Authority has to adapt itself not so
much to the people in general as to the actually existing institutions
which have the power to smooth or block its way. It therefore becomes
ideologically convenient to fall in with the general practice in the area
of identifying the existing agencies with the people, and permitting
de facto leadership in the region to be its own stamp of legitimacy. Thus
the grass-roots approach permits the Authority to come to terms with
one of the most important elements of potential resistance by approv-
ing the maintenance and strengthening of the existing administrative
apparatus at the state level. Moreover, since it may be assumed that
this apparatus and its control is the primary prize in the struggle
against federal encroachment, it is easy to see that the grass-roots ap-
proach permits a vital concession to the "states' rights" forces. All this,
once more, is quite apart from the conscious sources of the official
doctrine.
The attitude with which the TVA has had to contend may be indi-
cated by the following quotation from President James D. Hoskins of
the University of Tennessee, one of the "existing institutions" of the
Valley region:
... as Federal aid increases as it apparently must what will be the outcome in
the administrative field? Are we now to see the cherished democratic principle of
decentralization of authority and local control drowned beneath the flood of Fed-
eral dollars? Or, is there to be such a flood? Shall the Federal government operate
and control educational programs if state and local agencies cannot do so unaided?
Or shall the Federal government merely provide financial aid, and leave control to
the local authorities? 14
14 From President Hoskins' annual statement to the Board of Trustees of the Uni-
versity of Tennessee, Knoxville Journal, August 12, 1942.
62 University of California Publications in Culture and Society
President Hoskins went on to attack new trends in federal administra-
tion resulting in the by-passing of local agencies, particularly in the
field of agriculture. Yet TVA has received the public support of the
governors of the states in its region, particularly on the grounds of
satisfactory cooperation. 15 In this respect, that of coming to terms with
key institutions in its area, the TVA has used the grass-roots approach
to considerable advantage.
The grass-roots theory is not a mockery or fraud simply because it
has served a function tangential to democratic values. A living organi-
zation must use tools which are at hand, with due regard to consequences
for its own existence, accepting compromise as a normal state of affairs.
Such formulations as "concern for the people" remain, as doctrine, es-
sentially unanalyzed ; unless certain concrete procedures are specified,
such ideas are bound to be relegated to an organizational limbo, because
they will not be useful in embodying the principles in action. On the
other hand, once procedures are specified, then these procedures tend
to become the receptacle of the positive emotional values provided by
the general principle, even if what is put into practice is a theoretically
unsatisfactory rendering of the original moral imperative. In action
(and hence concretely), the method will be embodied in relations with
some actual institutions. These will of necessity share whatever moral
value may be attributed to the method ; as responsive structures, they
will invest the total relationship with a meaning determined by their
own origins and commitments.
The authority's staff is not unaware of this problem, although there
is practically no open criticism of the basic official doctrine. One long-
standing member of the staff felt that the leadership had betrayed its
real democratic responsibilities, and remarked :
The grass-roots policy is merely a rationalization. It is absurd to have a federal
agency trying to work through organizations which place it at a second remove
from the people. There is a distinction which must be drawn between "institutional
grass roots," represented by Senator McKellar, President Hoskins, and TVA's Agri-
cultural Relations Department, and a "popular grass roots" which would be less
concerned with the prerogatives of established leaders. Unfortunately, in practice,
the TVA has chosen the former.
This is an extreme view, but it serves to indicate the varying content
with which unanalyzed terms may be invested, and the way in which
the significance of the doctrine is bound up with its concretization.
15 See report of the Tennessee Valley governors on TVA in Hearings on S. 555
(Missouri Valley Authority), subcommittee of Senate Committee on Commerce, 79th
Cong., 1st sess. (April, 1945), pp. 34-38.
Selznick: TVA and the Grass Roots 63
Aside from the general problems of who are "the people," and the
need to make explicit the criteria of selection, there are other ambiguities
in the terms of the grass-roots doctrine :
1. There is a vagueness in the official doctrine about the mechanics by
which local influence may be brought to bear upon a decentralized
agency. It is well known that, where administrative control from the
center is weak, the field officers of an organization may tend to become
representatives of their area. As such, they are subject to the informal
influence of close association with local problems, including the pres-
sure of a local public which must be dealt with on a day-to-day basis.
An independent agency may indeed identify itself with the welfare of
its region, and shape its techniques of administration in local terms.
But it is not clear that the people of the Tennessee Valley have been
afforded any significant control over the operations of the TVA beyond
the normal channels of congressional appeal. The responsibility for
decisions of serious consequence lies with the Congress. The Tennessee
Valley, as a region, does not make decisions on such important matters
as the construction of dams or the expenditures of the regional authority.
2. It may be questioned whether the location of the central office of an
agency is decisive for the quality of being "close to the people." Long
association, involving acquaintance with field personnel as well as with
the habits, customs, and outlook of an agency, may be of equal impor-
tance. This is especially true if "the people" is identified with its
organized expressions, existing local governments and private associa-
tions. These institutions may build up a stable set of relationships with
agencies controlled from afar, and yet feel that they are close. The
defense of the Bureau of Eeclamation and the Corps of Engineers by
many local groups from the Missouri Valley, and their fear of some
new agency which, though decentralized to the area, might be alien,
is a good example of such a relation. The issue here is not the defense
of one approach against the other, but simply to point out that the
meaning of "close to the people" is unanalyzed.
3. The significance of "working through established agencies" may
vary, depending, for example, on the relative strength of the local
organizations. There may be effective cooperation among equals, or
dominance by one side or the other. Much may depend on the extent
of discretion and the degree of common agreement on general approach.
The development of administrative constituencies may significantly
shape the concrete meaning of the method. 18
10 See below, pp. 145 ff.
64 University of California Publications in Culture and Society
4. "Coordination" is another term whose meaning will be derived from
the procedures which are established for its effectuation. Like "unity,"
this is a word which is often used vaguely in order to avoid making
real objectives explicit. It is an honorific which must grace all admin-
istrative programs. But coordination involves control; and it is pre-
cisely the concrete lines of authority implied in any given system of
coordination which constitute its meaning. Coordination may range
from the mere circularization of ideas through the definition of juris-
diction all the way to the most thoroughgoing Gleichschaltung . In the
TVA, one leading official has made a distinction between the "coordina-
tion of agencies" as the function of a regional authority and the "coordi-
nation of programs," with the latter considered as proper. But of course
if this involves the elimination of other agencies, whose programs are
taken over by the regional authority, it will not be highly valued, as
coordination, by the agencies in question.
5. "Participation of the people" is rated highly ; but here again its
significance depends upon such factors as the meaning of the program
to the individuals concerned, the level of education, the actual locus
of decision, and the nature of the organizations through which partici-
pation is achieved. Participation may range from mere involvement by
means of devices established and controlled by the administrative
apparatus to spontaneous organization based on and integrated with
existing community patterns of cooperation. 17
ADMINISTRATIVE DISCRETION
The propaganda of the "outs," of those who need not take responsibility
for a definite course of action, may retain its unanalyzed elements with
impunity; but those who govern (in organizations large or small) are
forced to act, and to assume responsibility for the consequences of events.
As a leadership exercises its discretion in action, it generates a history
which serves to make concrete the unanalyzed abstractions of its
doctrine.
Discretion is a process of selection above all the selection of tools
for the execution of policy. But these tools are social facts with lives
of their own, with needs and demands, with determined responses which
do not permit an easy subordination to alien or indifferent ends. Tools
are recalcitrant; they ask a price, the price of commitment. He who
would induce cooperation must agree to be shaped in turn, to submit
to pressure upon policy and action. This give and take may not appear
17 See below, chap. vii.
SelznicJc: TVA and the Grass Roots 65
in its most significant forms in a contract or agreement. It may never
be verbalized.
The influence of recalcitrant tools upon policy may not be readily
apparent. It may not be directed to the specific program which prompted
cooperation. Since the consequences of cooperation ramify widely, a
cooperating group may execute a given policy initiated by another with-
out substantial deviation, but at the same time it may make demands in
another field as the price of continued cooperation. The meaning of the
relationship of author and agent, of initiator and tool, will not be under-
stood until the consequences of cooperation for decision making can
be traced, including the effect upon the position of the organization
among its contemporaries, and upon its nature in the light of long-run
objectives.
Management strives for discretion; a leadership seeks freedom of
movement, a range of choice, so that it may more ably invest its day-
to-day behavior with a long-run meaning. Prestige and survival are
not normally accepted as legitimate ends of administrative behavior,
especially in government and among the subordinate units of a larger
organization. But prestige and survival, even for the smallest unit (so
long as there are those who have a stake in it), are real factors in
decision, as all participants know. Lacking formal channels, necessarily
concerned with more than a substantive program, management requires
discretion as a means of introducing the factors of prestige and survival
into its choices among alternative methods for the execution of formal
policy. This factor is reinforced as the role of the managerial expert
grows in importance. Professionalization in management, providing
ideals of method and loyalty to managerial objectives, creates a "re-
sponsible management." But, as the price of responsibility, the expert
insists upon an enlargement of discretionary powers. He joins the
ranks of other experts in denying to the layman the right to judge
among alternatives when these lie within the province of specialized
experience.
Discretion is selection ; and it is also intervention. It enters upon an
existing social situation, creating disequalibrium and anxiety until
some new adjustment restores order and security. Intervention involves
consequences for (1) enduring factors always present where prior social
organization exists; and (2) conditions of imbalance or conflict which
happen to exist at the time of intervention.
1. Administrative intervention cannot escape the force of precedent.
Normally, the attempt to institute a program will run up against a
66 University of California Publications in Culture and Society
network of prerogative and privilege, which will have to be dealt with
in some manner before operations proceed very far. Except in extreme
cases, these interests will not be liquidated or even absorbed, but will
remain as centers of power which the organization must take into
account. This situation soon creates the problem, among others, of the
representation of interests. This may take explicit form, 18 or, when the
real pressures are strong and no formal avenues are permitted, repre-
sentation may be informal and sometimes covert. Indeed, where the
selective process implicit in discretion may invite criticism, the process
itself may become informal and covert. When there is a serious uneasi-
ness about informal representation, an ideology may be developed, or
an existing doctrine, perhaps otherwise neutral, may be reinforced. A
doctrine of legitimacy may be elaborated, justifying a specific selection
of representatives when alternative possibilities are present. Where a
program is broad, and representation of a general interest group or
class is at issue, choice by the agency often signifies intervention among
the competitors for leadership within the client group itself. This is a
decisive act of discretion, and may often commit an organization to a
faction with which it has no sympathy and to a conflict it does not
desire; or such discretion, when its consequences are recognized, may
be a means of modifying the social structure with which the organization
has to deal.
2. Discretion may occur in a context of contending social forces. The
loyalties of an administrative leader may be weaned away from exclu-
sive attention to the organization with which he is identified when the
interests of a social class or group are at stake in a general struggle. In
such a context, administrative decisions may be made in terms of con-
sequences for the over-all contest, with the needs of the particular
organization subordinated. An extreme case is the role of Communists
in labor unions, in particular their tendency to subordinate the inter-
ests of a union to the needs of the general defense and extension of the
policies of the regime in Soviet Russia. But the pattern is more general :
in a period of anticlerical struggle, it will be expected that a devout
Catholic administrator will devote some attention to the consequences
of his actions for the welfare of the Church ; in a period of considerable
ethnic tension, the decisions of a Negro or a Jew might be expected to
be shaped by considerations tangential to the special program of his
organization; when sectional feeling runs high, an administrator in a
18 See A very Leiserson, Administrative Eegulation : A Study in the Representation
of Interests (Chicago: University of Chicago Press, 1942).
Selznick : T VA and the Grass Boots 67
federal agency might subordinate his loyalty to the agency as such to
the struggle for the advancement of sectional goals.
The problem of administrative discretion is customarily raised in
terms of the potentially arbitrary action of quasi-legislative or quasi-
judicial regulatory bodies. 19 This is of great importance as the practice
of delegating authority to the executive power is extended. But from
the point of view of organizational analysis it is the internal relevance
of discretion which is especially significant. Administrators commonly
develop commitments to methods and to specific organizations with
which convenient relationships have been established; these in turn are
related to client groups with ends and commitments of their own. As a
consequence, a system of mutual dependence and aspiration develops
which cuts across organizational lines. Hence the exercise of discretion
must be analyzed with reference to the structure of the governmental
system as a whole, as well as to the evolution of the particular organiza-
tion itself. The execution of a program may be materially affected, in
the long run, by the managerial structures which are built and sus-
tained in connection with it. Insofar as discretion is permitted in the
building of any given organization and those related to it, a measure
of control over policy itself will in effect have been delegated.
The TVA's grass-roots doctrine seeks to maximize discretion. Depend-
ing upon the objectives of public management, this may be an un-
impeachable program. It is not the concern of this essay to offer an
appraisal. What is significant here is that the exercise of discretion
in a concrete social context and under the pressure of organizational
needs created a problem of interest representation, emphasized by
action within a situation marked by controversy among groups and
agencies. 20 In addition, the elaboration of a doctrine justifying the
maximization of discretion poses for the federal structure significant
alternatives unanticipated by the founders of the Authority. The wide
measure of corporate freedom afforded the TVA was related to its
business function and justified by the experimental character of the
project, as well as by the exigencies of the public power controversy.
This has been extended by the TVA itself to a program for the de-
centralized administration of regular federal functions. On this view,
the straightforward alternatives implicitly posed by the Authority for
19 See for example Harold J. Laski, "Discretionary Powers," Politico,, I, nos. 1-4
(1934-1935), pp. 274-285; also Marshall E. Dimock, "The Eole of Discretion in
Modern Administration," Frontiers of Public Administration (Chicago: University
of Chicago Press, 1936).
20 Discussed in chaps, iii-vi.
68 University of California Publications in Culture and Society
the future of the federal government are (1) no independent regional
agency; or (2) the subordination of departmental prerogatives to the
discretion of the regional agency. Since it is unlikely that the federal
departments or their organized clienteles will willingly abandon struc-
tures and precedents established after considerable effort, it is possible
that the concrete interpretation of grass-roots administration, evidenced
in the discretion exercised by TVA, may endanger the general idea of
decentralization itself: some of the older established agencies have
interpreted their experience with TVA to mean that an independent
regional agency cannot effectively coordinate federal agencies in the
field. The TVA leadership feels that it need not be concerned about the
consequences of its version of decentralization for the federal depart-
ments, for, it contends, a democratic perspective is concerned not with
the prerogatives of officials but with the needs and desires of the people.
But this view breaks down as we note that the special relation of these
federal agencies to client publics who have a stake in their continued
existence makes of them something more than simply bureaucratic
domains. Hence the problem of discretion assumes its full significance
only as we see its consequences for organizational character and relation-
ship, internal and external. 21
The organizational consequences of discretion are not often made
explicit. However, one TVA official, discussing the future of the Author-
ity's financing, made the following statement :
The most obvious result of comprehensive revenue-financing power is to give the
Authority largely independent discretion in the expenditure of funds on "develop-
ment activities." The miscellaneous activities falling within this category are the
ones spoken of in the statute with the least clarity. At the same time, they consti-
tute the integrating cement of the Authority's "regionalism." Also, they are the
activities most subject to conflict with the various activities of the various line
departments of the federal government, and of the Valley states. The content of
this development activities program has expanded and changed markedly in con-
junction with the evolution of the Authority's concept of an integrated regionalism.
With the completion of the major construction phase, the real character of the
Authority will depend upon the character of its development activities program. 22
The TVA does not have and has not sought the measure of discretion
involved here, but it is clear that a consideration of the implications of
discretion raises problems significantly related to the emergent character
of the organization itself.
21 For further discussion of the implications of the grass-roots policy for U.S.-TVA
relations, see below, pp. 74 ff . ; also chaps, v and vi.
22 Eef erenee here is to the possible financing of development activities, such as the
agricultural and regional studies program, out of power revenues, as these reach sur-
plus proportions.
Selznick : T VA and the Grass Roots 69
INHERENT DILEMMAS
Tension and dilemma are normal and anticipated corollaries of the
attempt to control human institutions in the light of an abstract
doctrine. Social structures are precipitants of behavior undertaken
in many directions and for many purposes. Mutual adaptation es-
tablishes only an uneasy equilibrium. This in turn is continuously
modified and disturbed as the consequences of action ramify in un-
anticipated ways. Practical leadership cannot long ignore the resistance
of social structure, and is often moved thereby to abandon concern for
abstract goals or ideals for which it is often criticized out of hand
by the moralists and idealists who lack experience with the vicissitudes
of practical action. 23 But a leadership which, for whatever reason, elects
to be identified with a doctrine and professes to use it in action, is con-
tinuously faced with tensions between the idea and the act. Ideological
symbols may fulfill useful functions of communication and defense and
may be long sustained as meaningful even when effective criteria of
judgment remain lacking; but an act entails responsibility, establishing
alliances and commitments which demand attention and deference.
This is not to suggest that ideals are futile and abstractions useless.
Tension does not mean defeat, nor does dilemma enforce paralysis. It
is precisely the problem of leadership to find a means, through com-
promise, restraint, and persuasion, to resolve tensions and escape di-
lemmas. But in doing so, attention must be directed to the real forces
and tendencies which underlie its difficulties. This is the constructive
function of analysis which seeks to take account of structural rigidities
and the indirect consequences of executive action. Where such analysis
is considered destructive, it is usually because doctrine, assuming an
ideological role, is not meant to be analyzed. In extreme cases, un-
analyzed doctrine ceases to operate in action at all, and the real criteria
of decision are hidden in a shadowland of unrecognized discretion de-
termined opportunistically by immediate exigency.
The TVA, in relation to its policy of grass-roots administration, is
not immune to such difficulties. Though seldom made explicit, sources
of tension are recognized by members of the staff, and have already
entered into the process of administrative decision. Among these may
be noted :
1. There is a dilemma of doctrine and commitment, or of the abstract
23 Cf . Kobert K. Merton, "The Role of the Intellectual in Public Bureaucracy."
Social Forces, XXIII (May, 1945), 413.
70 University of California Publications in Culture and Society
and concrete* This dilemma is the most general source of the tensions in-
herent in the grass-roots approach, for it inheres as well in all behavior
which involves both the verbalization of ideas and a set of specific activ-
ities. Doctrine, being abstract, is judiciously selective, and may be
qualified at will in discourse, subject only to the restrictions of sense
and logic. But action is concrete, generating consequences which define
a sphere of interest and responsibility, together with a corresponding
chain of commitments. Fundamentally, the discrepancy between doc-
trine and commitment arises from the essential distinction between the
interrelation of ideas and the interaction of phenomena. The former is
involved in doctrine, the latter in action. Of course this is the ground
of the normal necessity to revise decisions and even over-all doctrine
in the light of events. The tension between the abstract and the con-
crete is resolved through continuous executive action. However, where
doctrine itself creates commitment, as in the institutionalization of
policy, executive decision is not readily reversible. Policy which osten-
sibly should be determined on the basis of a scientific appraisal of prac-
tical means for the achievement of formal ends becomes invested with
prestige and survival value and may persist as official doctrine despite
a weakening of its instrumental power. Whatever instrumental capacity
the policy does have is related to informal rather than professed goals.
2. There is also the dilemma of consent and conformance 25 or of selec-
tive decision and total involvement. Democracy as method, and the
grass-roots policy as method represent processes of decision. Decision,
however, demands only the partial consent of the participants, who are
involved only obliquely in their capacity as voters or choosers. But the
execution of decision is a matter of action, which tends to involve the
participants as wholes. Hence cooperative action, as Barnard says,
"requires substantially complete conformance."
In other words, while the choice of a given course of action may be
conceived of as involving the individual or group only to a limited
degree, in fact there is a tendency for circumstances to demand more
extensive involvement. Organizational action, once initiated, tends to
push onward, so that the initiator may be enmeshed in new relation-
ships and demands beyond his original intention. Here again the key
word is "commitment." Every executive knows that the initiation of a
new course of action is a serious matter precisely because of the risk
involved that the establishment of precedents, of new machinery and
24 See C. I. Barnard, Dilemmas of Leadership in the Democratic Process, Stafford
Little Lectures (Princeton: Princeton University Press, 1939), p. 11.
Selznick : T VA and the Grass Roots 71
new relationships, the generation of new and complex interests may
make greater demands upon his organization than he can presently
foresee. The problem is not one of inner impulse, but rather of the
structural forces which summon action and constrain decision. This
structurally induced tendency is roughly comparable to the notion of
adience as applied in the theory of animal drive.
Conf ormance, indispensable to the completion of the movement begun
by decision, has therefore a qualifying effect and creates an inescapable
tension. This means, in respect to the TVA, that while its decentralist
policy may be instituted for special reasons (referrable in part to the
ideals of the leadership as well as to other factors) , nevertheless there
will be a tendency for the organization as a whole to be shaped by the
process of conformance. It will have to go farther in carrying out the
policy than it may have originally intended; the formal process of
consent will have been transformed into concrete institutional relation-
ships, generating new and unlooked-for demands. Generally, the prob-
lem of checking conformance is present in all administrative decisions
concerning relationships established under the grass-roots formula.
3. So far as discretion is delegated, bifurcation of policy and admin-
istration is reinforced. It is basic policy inside the Authority that plan-
ning and execution should be united in a single administrative organ.
But the delegation of functions involved in the grass-roots approach
makes insistence upon this principle of unity somewhat anomalous. The
channeling of programs through independent agencies necessarily dele-
gates discretion, hence programs may be extensively modified in execu-
tion. The dilemma is only made more explicit if controls are instituted
which operate objectively to transform the independent local agency
into an administrative arm of the Authority : execution may be brought
into line with policy, but the grass-roots objective will have been under-
mined. The dilemma is mitigated to the extent that a denial is made of
the possibility of a difference in objective between the initiating and
the executing agency. An attempt in this direction made in the TVA's
agricultural program will be described in a later chapter.
4. Theories about government become preempted by social forces. It
is unrealistic to estimate the implications of a theory of government on
the basis of its abstract formulation. The propagation of a point of
view carries with it an often unwished-for alliance with others who,
for their own reasons, are espousing the same or a convergent doctrine.
A theory about method may at any given time be linked with a special
substantive doctrine; support for the method many imply a position on
72 University of California Publications in Culture and Society
program, and conversely. Thus TVA finds itself in the camp of the sup-
porters of "states' rights" both in its criticism of over-centralization in
government and in its support of the state governments as "regional
resources" in its area of operation. But at the same time, in respect
to the complex of political issues summed up in the extension of "positive
government," the Authority's point of view is very far from that of
the general run of supporters of local sovereignty. TVA is therefore
in the continuously ambivalent position of choosing between an em-
phasis on method and an emphasis on substantive program, or more ac-
curately, of assigning a priority to one or the other. To the extent that
the movement for planning and a strong federal government oriented
along welfare lines is identified with the existing federal structure, the
TVA leadership tends to emphasize method as basic and cut itself off
from the general welfare movement. 28 But it cannot divorce itself com-
pletely from its antecedents, and the ambivalence persists.
5. Emphasis on existing institutions as democratic instruments may
wed the agency to the status quo. A procedure which channels the admin-
istration of a program through established local institutions, govern-
mental or private, tends to reinforce the legitimacy of the existing
leadership. This is especially true when a settled pattern claims the exclu-
sive attention of the agency, so that other groups striving for leadership
may find their position relatively weakened after the new relationships
have been defined. In strengthening the land-grant colleges in its area,
the TVA has bolstered the position of the existing farm leadership. There
is some evidence that in the process of establishing its pattern of coopera-
tion, TVA refrained from strengthening independent colleges in the
area not associated with the land-grant college system. Again, the rela-
tively dominant role of the American Federation of Labor unions in
TVA labor relations, especially as constituting the Tennessee Valley
Trades and Labor Council, is objectively a hindrance to the develop-
ment of labor groups having other affiliations. In general, to the extent
that the agency selects one set of institutions within a given field as
the group through which it will work, the possibility of freezing existing
social relationships is enhanced. At least in its agricultural program,
TVA has chosen to limit its cooperative relationships to a special group,
so that the potential or inherent dilemma has been made explicit.
28 This is not yet clear to all participants, because the public power controversy is
still important, and allegiance to TVA is denned in terms of its role in advancing
public power. Support of the authority method in other areas derives much of its
strength from the same sources which originally shaped the structure of TVA : the
avant-garde quality in terms of national politics and the need for flexibility in carry-
ing forward the electric power controversy.
Selznick : T VA and the Grass Roots 73
6. Decision at the grass roots may be inhibited by the system of na-
tional pressure groups. When important issues become crystallized in
the program of a group organized on a broad scale with a national lead-
ership, local decisions may be influenced primarily by their effect on
the outcome of over-all controversy. The local problem is appraised not
for its own sake but for the influence of a local decision on the general
bargaining position of the leadership. Thus a local branch of the Amer-
ican Farm Bureau Federation or of a CIO union may have its attitudes
framed for it by a long-run national strategy, as by a pending bill in
Congress for the transfer of functions from one agency to another, or
an impending organizing drive, or a national election. To the extent
that a decentralized governmental agency is influenced by decisions
made by local groups in national terms, it would appear that the grass-
roots approach becomes the victim of the growing centralization of
political decision.
7. Commitment to existing agencies may shape and inhibit policy in
unanticipated ways. When the channels of action are restricted, pro-
grams may be elaborated only within the limits established by the
nature of the cooperating organizations. The traditions and outlook of
an established institution will resist goals which appear to be alien,
and the initiating agency will tend to avoid difficulties by restricting
its own proposals to those which can be feasibly carried out by the
grass-roots organization. Where the grass-roots method is ignored, new
institutions may be built, shaped ab initio in terms of the desired pro-
gram. An attempt to carry forward a policy of nondiscrimination (as
against Negroes) will not proceed very far when the instrument for
carrying out this policy usually as an adjunct of some broader pro-
gram has traditions of its own of a contrary bent. Moreover, the grass-
roots policy voluntarily creates nucleuses of power which may be used
for the furtherance of interests outside the system of cooperation orig-
inally established. Thus the TVA distributes electric power through
electric power boards which are creatures of municipalities, with the
contractual reservation that surplus income shall be used only for im-
provements in the system or for the reduction of rates. But the question
has been raised: what if pressure arises to use surpluses for general
purposes, that is, to finance nonpower functions of the municipal gov-
ernments? And what if the state governments undertake to tax these
surpluses, because of a restricted tax base and unwillingness to institute
a state income tax? The logic of the grass-roots policy might force the
Authority to agree. However, it is perhaps more likely that the Author-
74 University of California Publications in Culture and Society
ity's commitment to function as a successful power project would take
precedence over the grass-roots method.
8. Existing agencies inhibit a direct approach to the local citizenry.
The participation of local people always takes place through some or-
ganizational mechanism, notably voluntary associations established to
involve a public in some measure of decision at the end-point of opera-
tion. But such associations are commonly adjuncts of an administrative
agency which jealously guards all approaches to its clientele. If, there-
fore, a federal agency establishes cooperative relations with such an
agency, it will be committed as well to the system of voluntary associa-
tions which has been established. Hence the channels of participation
of local people in the federal program will be shaped by the intermediary
agency. In respect to its closeness to the people, the status of the fed-
eral government may not, in such circumstances, be materially altered.
Viewed from this perspective, the grass-roots method becomes an effec-
tive means whereby an intrenched bureaucracy protects its clientele,
and also itself, from the encroachments of the federal government.
IMPLICATIONS FOR U. S.-TVA RELATIONS
The grass-roots theory has served the special needs of the TVA as an
organization, above all the need for acceptance in its area of operation.
It is not surprising, therefore, that the idea has not been wholeheartedly
welcomed by agencies of the federal government whose responsibilities
are national rather than regional; and that the TVA has chosen to
clash head on with some Washington departments, thus failing to make
an adjustment to that part of its environment constituted by the existing
structure of the federal government. In the early days of TVA, the so-
called "New Deal" agencies were unwilling to interfere with a social
experiment that was an important symbol in- the over-all political strug-
gle. But as TVA's position became secure, such organizations as the
U. S. Department of Interior began to consider the implications of the
idea of an independent regional authority for their own future. Doubts
were transformed into urgencies as the movement for the extension of
the TVA approach to other regions gained headway.
The extension of the TVA administrative structure, especially its
managerial autonomy, to other areas, poses far-reaching alternatives
for the future of the federal government. If broad responsibilities for
regional development are lodged in a single agency, and such agencies
are multiplied so that they include virtually the entire area of the
nation, the operational functions of the developmental agencies in
Selznick : T VA and the Grass Roots 75
Washington may be substantially curtailed or even eliminated. Driven
to a logical (if somewhat farfetched) conclusion, such agencies would
primarily assume staff functions, involving a radical bifurcation of
planning and execution. At the other extreme, regional agencies would
themselves be purely advisory bodies or operating arms under the con-
trol of a federal department, or would not exist at all. These long-run
implications have not been lost to the leading participants in Wash-
ington and Knoxville.
For the TVA the question of the structure of future regional author-
ities is not academic. Partly it involves the prestige of the leadership,
because of personal and professional commitment to an embattled
method. 27 But perhaps more urgent is the realization that TVA itself
may not long maintain its autonomy if a national decision is made
against the autonomy of other regional agencies. The Authority would
then become an anomaly, and in time would be coordinated with the
structure as a whole.
"If there is one thing," Lilienthal has said, "which the nine years of
TVA experience have proved to us, it is this: unity of management
located physically in the region is a sine qua non in getting results."
With this conviction as a foundation, he has argued repeatedly that
the establishment of regional development authorities cannot be viewed
apart from the question of administrative structure, insisting that the
TVA pattern of autonomy is essential to successful operation. This
autonomy includes such minimum essentials as freedom from veto by
such control agencies as the General Accounting Office ; freedom from
control by the U. S. Civil Service Commission; freedom to use current
revenues for operating purposes ; and unified control over its own de-
velopmental programs so that the policy of cooperative administration
with and through local and state agencies may be effectuated. The fol-
lowing informal statement by Lilienthal indicates the quality of urgency
with which the debate (at least from the TVA side) is suffused:
We have observed a tendency among those who wrestle with this problem (of develop-
ing natural resources on a regional basis) to assume that administrative arrange-
ments in dealing with regional development problems are relatively unimportant.
I cannot emphasize too strongly that our experience shows that quite the contrary
is true. The things that have happened in the Tennessee Valley, for which the TVA
is in any way responsible, would not have happened but for the fact that there is
nothing vague, or "fuzzy," or indeterminate about the central administrative respon-
27 That this type of commitment is important is supported by the assertion of one
TVA official that a leading proponent of the extension of civil service control cannot
understand the TVA viewpoint on an autonomous agency civil service because of his
long-standing professional commitment to the U. S. Civil Service Commission.
76 University of California Publications in Culture and Society
sibility and commensurate authority and autonomy which the TVA carries under the
law. Cooperative arrangements, devices of coordination, and the fullest collaboration
among federal and state agencies are characteristic of the TVA's methods of getting
the job done, but these methods would go for naught, in our judgment, unless there
were a central management able to make important decisions and to provide a core
of leadership in the region, reasonable autonomy, and independence from minute
direction or delays of paralytic indecision from Washington.
The issue, Lilienthal suggests, may be difficult to resolve if the preroga-
tives of the Washington bureaus must be taken into account, but it is
simply answered if "the people, the public, have to be satisfied." But
just these prerogatives are of vital importance to the existing federal
agencies : hence the conclusion that TVA has elected not to make the
same adjustment to the existing pattern of national government as it
has to the existing pattern of state and local government. The grass-roots
doctrine is a vehicle for the accomplishment of the latter ; at the same
time it is an instrument of disaffection from the Washington establish-
ments. It is, of course, not the purpose of this discussion to enter the
debate in question, or to make any appraisal of relative merits. We are
interested here only in the question of the implications of the grass-roots
doctrine for the position of the TVA as an organization.
The self -consciousness of administrative agencies is a factor in the
behavior of all leading participants. A formulated doctrine is a prop to
self-consciousness and a vehicle for its expression. This assumption
is important for the interpretation of the grass-roots theory as a means
of defining the approach of TVA to other organizations. The role of the
doctrine becomes significant as it informs the handling of current issues,
as it invests them with a long-run meaning. This is true of all organiza-
tions, though some are more prone to verbalize than others. An example
in the TVA, which highlights the potential strains in its relations with a
federal agency, may serve to make this clear. In 1941, the Department
of Agriculture established certain quotas for increased agricultural
production, channeling its program through the State USDA Defense
Boards and State Land-Use Planning Committees, which included rep-
resentatives of the state extension services of the land-grant colleges.
This procedure was followed in the Tennessee Valley without clearance
through the Correlating Committee jointly representing the TVA, the
USDA, and the Valley states land-grant colleges. Presumably, the De-
partment felt that its responsibilities under the defense program could
best be carried out through organizations which it, independently of
relations to the TVA, had established on the state and local level. The
Department's program for Tennessee called for the expansion of quotas
Selznick : T VA and the Grass Roots 77
for feed corn, which involved what it considered a small net increase in
clean-tilled acreage. The TVA took issue with the Department on this
matter, suggesting that the desired increase in livestock and livestock
products could better be obtained by expanding the acreage, production,
and yield of pastures and meadows, thus continuing the conservation
policies, supported by the Authority, of reducing row crops and aiding
in the development of cover crops by the distribution of phosphatic
fertilizers. The Department maintained its viewpoint, insisting that it
was coordinating conservation farming with the national interest in
increased production through the agency of state agricultural leaders
who had included considerations of erosion and related conservation
factors in their determination of local plans and quotas.
This difference in approach seemed to the TVA a significant example
of inadequate definition of national policy in regional terms. The Author-
ity, of course, could do nothing about it, but its interpretation of the
incident in self-conscious and long-range terms is significant. One lead-
ing TVA official pointed out that he felt there was a real danger that
the Department, in expanding food production, might undertake activi-
ties inconsistent with the program in the Valley, which integrated crop
production with the regional problem of land and water control. More-
over, he pointed out, "the current trend of deciding all such matters
at Washington, without the careful consideration of special local cir-
cumstances which the TVA and cooperating agricultural colleges can
supply, is a threat to the long-time effectiveness of regional organizations
such as TVA." (Emphasis supplied.) Thus, the Department's involve-
ment of state and local agricultural committees, presumably a grass-
roots approach, is discounted, and the implications for TVA as the
integrating regional agency has come to the fore. This official stressed
the significance of the issue for the long-time destiny of TVA, and
pointed out that in the postemergency programs TVA should "fight to
prevent the entire regional approach to national problems from being
submerged by the pressure of emergency needs, and by the dangerous
tendencies toward overcentralization of public policy determined in
Washington." In this way, the grass-roots doctrine, when identified
with the indispensibility of TVA, serves to give meaning to current
problems, so that they are related to long-time organizational strategy.
It is interesting that the TVA program is referred to as the "Tennessee
Valley region's program of land and water resource development" in-
volving a somewhat gratuitous identification of the TVA, as an agency,
with the region itself. The USDA program, on the other hand, is con-
78 University of California Publications in Culture and Society
sidered as divorced from the region, despite its cooperation with many
of the same local organizations with which the TVA works. In this case,
with respect to federal agencies, the grass-roots doctrine functions pri-
marily to defend the TVA's autonomy and its central status with respect
to the region.
The point of view of the Washington agencies, rejecting the idea of
autonomy for regional authorities, has been made clear as proposals
for the extension of the TVA idea have come up for consideration.
Several of these points of resistance to the TVA approach on the part
of federal agencies may be adduced :
1. In 1941 a bill was introduced in Congress which proposed to include
the Cumberland River and its basin within the provisions of the TVA
Act of 1933, thus widening the jurisdiction of the Authority to a con-
tiguous river basin. This measure was opposed by the Department of
Agriculture, which prepared a report outlining its views for the Com-
merce Committee of the U. S. Senate. It was pointed out that the TVA
has been made responsible in its region for activities within the field
of agriculture and conservation for which the Department of Agricul-
ture had a nation-wide responsibility. Because of the TVA's program,
the Department had not extended its erosion control work to the Ten-
nessee Valley. However, even though satisfactory arrangements had
been made with the TVA to avoid duplication, it was felt that the prin-
ciple involved was not sound and should not be extended. The TVA
arrangement, the Department held, was tolerable as an exception, but
could not be accepted as a rule. The Department expressed its apprecia-
tion of the need for planning in the development and conservation of
natural resources, and favored the establishment of regional planning
agencies which, however, would have authority only to recommend
projects to other agencies, Congress, and the President, but would not
have authority to carry them out on their own account. Thus the TVA
approach was rejected in principle, posing for regional agencies the
same logical alternative which TVA implicitly posed for the federal
departments a staff or advisory status.
2. In a preliminary report on the Arkansas Valley, issued in 1942 by
the National Resources Planning Board, including a comprehensive eco-
nomic and social plan for the region, the administrative recommenda-
tions failed to follow the TVA pattern. Among the 45 persons listed as
having made material contributions to the preparation of the plan there
were:
7 from the Bureau of Agricultural Economics, USD A
Selznick: TVA and the Grass Roots 79
4 from the Soil Conservation Service, USDA
4 from the Farm Security Administration, USDA
2 from the Forest Service, USDA
7 from the National Resources Planning Board
2 from the Oklahoma College of Agriculture
2 from the University of Arkansas
1 from the University of Texas
1 from the Southwest Cooperatives
2 from the Rural Electrification Administration
The Reclamation Service, Corps of Engineers, and other federal bu-
reaus were actually to execute the program. It is not surprising that
these agency-minded men did not vote a lack of confidence in their own
organizations. But it is significant that the NRPB's planning was (and
had to be) influenced by the existing departmental structure. Even if the
voice of TVA were raised in its councils, it could scarcely be expected
to carry much weight. In the minds of the TVA officials, the NRPB
was, as one of them put it, "preparing to butcher the TVA idea" in its
report. In any case, it was clear that even an organization like NRPB,
in the Executive Office of the President and separate from the operating
departments, could not escape the federal vortex. This does not mean
that the NRPB as an organization was opposed to the idea of regional
decentralization of federal functions, or even to the authority method
as such. It does mean that a federal planning agency with national re-
sponsibilities, leaning upon existing federal agencies for cooperation,
could not embrace a doctrine which might be interpreted as basically
inimical to the long-time interests of established organizations within
the federal structure. The TVA leadership does not appear to have used
the same critera of realism in approaching the possible response of the
federal bureaus as it has in taking account of the interests and preroga-
tives of state and local agencies. This discrepancy seems to provide a
warrant for the conclusion that TVA found adjustment to its area of
operation a prior necessity; in making that adjustment, it committed
itself to a theory which could not long be maintained without tension
and conflict.
3. The undercurrent of debate and disaffection which for some years
had characterized TVA's relations with a portion of the federal structure
became most clearly evident in 1945, in the course of public hearings on
a proposed Missouri Valley Authority. 28 The problem of administrative
28 See the two sets of Hearings on S. 555, "A Bill to Create a Missouri Valley
Authority," Senate Committee on Commerce and Senate Committee on Irrigation and
Eeclamation, 79th Cong., 1st sess. (1945).
80 University of California Publications in Culture and Society
method was clearly highlighted because an extensive program of flood
control, irrigation, and power development had already been author-
ized by Congress, placing responsibility for execution in the Corps of
Engineers and the Bureau of Reclamation. In respect to this program,
therefore, the problem was not what was to be done so much as by what
means to do it. The Bill under discussion provided for an authority
which might carry out the water control program on its own account,
as well as make plans and initiate projects for the economic and social
development of the region. The MVA was to be modeled after the TVA
in its structure, with the same basic autonomy.
Among government officials, the chief antagonist of an autonomous
MVA was Harold L. Ickes, then Secretary of the Interior. While he did
not oppose the idea of regional authorities responsible for over-all con-
servation and development, Ickes wished to obviate the threat to the
existing departmental structure implicit in the extension of the TVA
pattern. His objections appeared to be based on the experience of the
Department of Interior with the TVA, as indicated in the following
statement :
. . . almost without exception, the assistance rendered TVA by the bureaus of In-
terior, and also, I might add, by bureaus of the Department of Agriculture, was not
always accepted by TVA happily. At times, there has seemed to be in TVA a feeling
that it was all-seeing, all-knowing, and all-providing within its valley, and that other
government agencies were interlopers, even when they sought no more than to pro-
vide customary Federal services; at other times unfortunate ventures in untried
fields resulted in the acceptance by TVA of preferred assistance, notwithstanding
prior rejections. Fortunately, on the whole, that hard taskmaster experience seems
to have convinced TVA, after it gradually lost the first exuberance of youth, that
success could sometimes better and more economically be achieved through existing
agencies of government. Yet, even today, there are some fields of endeavor where
needless and expensive overlappings exist between TVA and other governmental
agencies.
It is not too much to say that one of the most important lessons which we have
learned from the TVA experiment is that time, money, and manpower can be saved
if future authority legislation incorporates some element of compulsion to insure
efficient collaboration between authorities and the rest of the Federal government. 20
Ickes deprecated the idea that TVA's autonomy was the primary ele-
ment in its achievements, stressing the aid provided by the Public Works
Administration and the Bureau of Reclamation in the early days of the
Authority. The former arranged for considerable financial support dur-
ing a period when the constitutionality (and hence the very existence)
of TVA hung in the balance. "It was this support," the Secretary testi-
29 Searings on S. 555, Senate Committee on Commerce (April 18, 1945), pp. 122 ff.
Selznick : T VA and the Grass Roots 81
fied, "that made it possible for TVA to go forward with its power pro-
gram and become what it is today." Yet despite the claim of TVA that it
cooperates fully with federal agencies, Ickes insisted that the TVA pat-
tern in other regions would create a "constant pressure for exclusion"
of older organizations.
For this analysis, the significance of Ickes's attack on TVA (its merits
are not here considered) is that TVA's insistence upon autonomy, but-
tressed by the general doctrine of grass-roots administration, militated
against an effective adjustment to the federal administrative structure.
Thus TVA's insistence 30 upon its own terms if the authority method is
to be extended, tends to bolster political opposition to all regional devel-
opment agencies, autonomous or otherwise. This becomes significant as
it is recalled that, in the context of national controversy over social
issues, TVA remains a political symbol.
The MVA debate emphasized what appears to be the anomalous posi-
tion of TVA in its insistence on local autonomy as essential to the author-
ity method. Whatever its own desires, TVA cannot help being caught
up in the general struggle over the extension of positive government.
In that struggle, some federal agencies, especially those inaugurated or
expanded under the New Deal, are pitted against forces which oppose
planning and paternalism. The latter tend to be associated with states'
rights and antibureaucratic slogans, with which the TVA grass-roots
doctrine seems to converge. But that convergence cannot erase the fact
that the primary meaning of TVA as a symbol is not its administrative
structure but its role as a "microcosm of sovereignty," with the broadest
powers of the federal government for resource development and the
extension of the general welfare. The TVA idea, in the eyes of its lead-
ership may more and more come to be identified with a method of admin-
istration, but where issues are vital no administrative sleight-of-hand
will serve to obscure the real meaning of the social symbol. This was
made clear by Will M. Whittington, U. S. Representative from Missis-
sippi, among others, in testimony at the MVA hearings. Whittington
asserted his opposition to the authority or regional plan, and began his
list of reasons with the following :
1. Authorities are advocated by the spenders and by those advocating projects
unable to stand on their own merits. They are advocated by the planners and by the
planning agencies. They are advocated by those who believe in reforming and remak-
ing America. They are advocated by those who believe that the Government should
do for the citizens what the citizens should do for themselves. There is a place for
30 Including a judicious measure of lobbying, by no means, of course, unique to
TVA among governmental agencies.
82 University of California Publications in Culture and Society
public improvements. But there is a place for private initiative. There is no place
for regimentation.
2. Authorities contemplate uncontrolled bureaucracy that leads to irresponsible
bureaucracy. Authorities are the dream of bureaucrats. They involve the establish-
ment of a super-agency responsible to no one but itself. 31
With the issue defined in these terms, the protagonists of the TVA
method might be expected to subordinate questions of administrative
structure to unity on the central points in controversy, the concepts of
positive government and regional planning. But it is likely that the TVA
leadership, or any similar leadership, would find small comfort in a
victory for the principle of regional planning at the cost of its own ad-
ministrative autonomy. At the same time, the proponents of that prin-
ciple outside the government tend to lend their support to whatever
specific measure is currently an issue, if it will advance political fron-
tiers. Neither TVA nor the other interested federal agencies will make
decisions in purely political terms, for unlike the nongovernmental
groups, they have a stake in the consequences of specific measures for
particular administrative organizations. Structures embody interests,
which demand that leadership take heed of present doubts and future
threats. In this way, the administrative leaders may well be disqualified
from participation in a single-minded pursuit of political goals.
31 Searings on S. 555, Senate Committee on Commerce (April 23, 1945), p. 344.
f
Part Two
THE INFORMAL COOPTATION OF GRASS-ROOTS
ELEMENTS AN ADMINISTRATIVE
CONSTITUENCY AND ITS CONSEQUENCES
CHAPTER III
f
TVA AND THE FARM LEADERSHIP : THE CONSTRUCTION
OF AN ADMINISTRATIVE CONSTITUENCY
In adhering to the memorandum of understanding, the Authority is not protecting
simply an institution but also the principle of our concept of democracy. If there
had been no land-grant colleges in existence in 1933, in order to carry out an inte-
grated program in the Tennessee Valley region the Authority would have found it
necessary to recommend and support the establishment of such institutions.
DIRECTOR OF AGRICULTURAL RELATIONS,
TENNESSEE VALLEY AUTHORITY (1941)
THE PRECEDING chapter has been devoted to an interpretation of the
grass-roots idea as official TVA doctrine. Such an analysis may stand
by itself in some respects. But if we ask for the meaning of a policy in
terms of its consequences in action, we must trace the history of the
events within which the policy is presumed to have been effective. More-
over, it is impossible otherwise to document the hypothesis suggested
above that the grass-roots policy functioned for TVA as a mechanism
by which the agency achieved an adjustment to institutional forces
within its area of operation.
Among the varied efforts of the TVA to execute its responsibilities
under the guiding principle of grass-roots administration, the Author-
ity's agricultural program is probably the outstanding example. This
activity is a major phase of the TVA's work, and accounts for most of
the funds expended under contractual arrangements with local agencies. 1
The agricultural officials within the Authority constitute the most vig-
orous proponents of the grass-roots approach, and it is in this field that
the possible implications of the administrative principle have been most
clearly worked out in action. 2 We therefore turn to this major phase of
TVA's program as a case study of the policy of executing a regional
program with and through existing local institutions.
TVA'S AGRICULTURAL RESPONSIBILITIES
The TVA agricultural program derives in theory from the idea of a
unified development of the watershed of the Tennessee River, whereby
water control on the land complements the control of water in the river
1 See table II, p. 101 below.
2 It should perhaps be said that a number of officials within TVA would consider
the agricultural program a somewhat extreme rendering of the grass-roots method.
However, the central place afforded to it here is in fact a reflection of its equally
[85]
86 University of California Publications in Culture and Society
channel. With an average rainfall in the Valley of 52 inches, rising in
some areas to 85 inches, a long history of erosion, and a farm economy
which has emphasized the production of clean-tilled crops such as corn
and cotton, the TVA links its agricultural activities to over-all responsi-
bilities for the protection of the watershed through the storage of water
in the land.
In practical terms, however, it was the existence of the nitrate plants
at Muscle Shoals, Alabama, which determined the primary content of
the Authority's agricultural program. These plants were erected during
the First World War and in line with provisions of the National De-
fense Act of 1916 various efforts were later made, without success, to
put them to peacetime use for the benefit of agriculture. Though the
potentialities of the installations for commercial operation were subject
to prolonged debate, pressure for the utilization of Muscle Shoals as a
fertilizer project was persistent. In 1921, the significance of Muscle
Shoals was formulated by the American Farm Bureau Federation :
Nitrate Plant No. 1, Nitrate Plant No. 2, and the Wilson Dam together constitute
what is probably the greatest single conservation activity of our government. This
entire project should be viewed in the same light as an irrigation project, a forest
reclamation activity, or a levee or drainage problem. Its great purpose in peace
times, was to assist in maintaining our soil fertility and consequently in the adequate
protection of food for our increasing millions. 3
The reports of President Coolidge's Muscle Shoals Inquiry (1925) and
President Hoover's Muscle Shoals Commission (1931) lent support to
the drive for quantity production of fertilizers at the Shoals, although
there was little response from private industry for operation of the
plants. Efforts made to lease the properties on the basis of commitments
for large-scale production of nitrogenous fertilizers were unsuccessful.
Moreover, the chief supporter of government operation, Senator George
W. Norris, had little confidence in fertilizer potentialities of the project. 4
important role within TVA itself. Even were that not so, the selection of an extreme
example is justifiable on theoretical grounds. It is precisely the extreme case, in
which counteracting effects have been minimized, which permits an examination of
the principle laid bare.
8 Eeported in a brief of the American Farm Bureau Federation submitted to the
Congressional Joint Committee to Investigate the Tennessee Valley Authority, Nov.
21, 1938 (mim.), p. 2. Emphasis in the original. The good faith of the American
Farm Bureau Federation with regard to the fertilizer problem has been called in
question at various times ; but this does not vitiate the symptomatic significance of
the Federation's demands for fertilizer production. Moreover, the national Farm
Bureau's active support of the Soil Fertility Bill (See Hearings on S. 1251, Senate
Committee on Agriculture and Forestry, 80th Cong., 1st sess., May, 1947), against
the opposition of the fertilizer companies, operates to discount such criticism.
4 See C. Herman Pritchett, The Tennessee Valley Authority (Chapel Hill: Univer-
sity of North Carolina Press, 1943), pp. 9-13.
Selznick: TVA and the Grass Roots 87
It soon appeared that the real prize at the Shoals, both for those who
advocated government operation and for the private companies whose
bids were considered, was the electric power generated at Wilson Dam.
Nevertheless, the fertilizer aspect played an important role in the con-
troversy up to the last. Bids of the Ford Motor Co. and the American
Cyanamide Co. both involved the large-scale production of mixed ferti-
lizers. These bids were supported by the American Farm Bureau Fed-
eration, 5 but much of the argument against them was based on the theory
that the companies were in reality primarily interested in the available
power and would not give adequate attention to the fertilizer program.
The point is that the Muscle Shoals problem revolved around fertilizer
as a major focal point.
Pressure for large-scale production of fertilizers continued up to
within a year of the drafting of the TVA Bill. Although the earlier
emphasis of the Farm Bureau on the need for nitrates was modified as
prices for these products fell in the postwar period, its public position
on the Muscle Shoals issue was maintained. Thus, in 1932, Edward A.
O'Neal, President of the American Farm Bureau Federation, after
pointing out that the price of nitrogen had fallen, stated :
. . . but we must look to the future to protect our American farmers, who use so
much nitrogen, against a price for this product which existed formerly. Nitrogen per
pound is still the most costly of the three main plant-food requirements. ... I have
not surrendered the thought which was advocated ten years ago relative to the
importance of Muscle Shoals along nitrogen fixation lines. I merely want to convey
to the Military Affairs Committee the additional thought that this phosphoric-acid
proposition, particularly the prospect of developing phosphoric acid by electrolytic
processes is an additional possibility at Muscle Shoals to that which was advocated
formerly. 6
At the same hearings, O'Neal emphasized : "We didn't want any experi-
mentation in fertilizer. We want fertilizer as a dominant factor." And
on the same day a Southern Congressman called for the production at
the Shoals of enough fertilizer to control the national price.
In the light of these definite statements made as late as 1932, and
during the entire period of the controversy after 1920, the attitude of
the Farm Bureau toward the TVA is not altogether understandable.
After the election of President Roosevelt, this powerful lobby swung to
the support of the Norris Bill, resulting in a double compromise of its
3 For the record of the Farm Bureau's positions on these offers and related matters,
see Hearings, House Committee on Military Affairs, 72d Cong., 1st sess. (January 7,
1932), pp. 294-298.
8 Muscle Shoals Hearings, House Committee on Military Affairs, 72d Cong., 1st
sess. (January, 1932), p. 266.
88 University of California Publications in Culture and Society
earlier position : it no longer opposed a public corporation at the Shoals
and now accepted a statute which gave no guarantees as to quantity
production of fertilizers. 7 It is possible that an over-all agreement with
the new administration in Washington included a commitment by the
Farm Bureau to support the resolution of the Muscle Shoals controversy
on the terms of President Roosevelt. He was interested primarily in the
public power aspect of the development and in the possibilities of re-
gional planning, but apparently not in waging a price war against the
fertilizer manufacturers. 8 At the hearings on the TVA Bill, the Farm
Bureau supported the New Deal measure. Significantly, the Farm Bu-
reau did not request a definite commitment from the new agency for a
minimum production of fertilizer, either nitrates or phosphates, al-
though its representative, O'Neal, was pressed to do so by a member
of the committee. 9
It is this history of controversy over the fertilizer issue, plus the prior
existence of manufacturing installations at the Shoals, which has cen-
tered the TVA agricultural program upon the production and distribu-
tion of fertilizers. However, the TVA activity in this field has been
defined as experimental and has in no sense fulfilled the initial demand
for large-scale production.
Although fertilizer is the central axis of the agricultural program,
that program has a more general relation to the objectives and problems
of the Authority. The need to hold the water on the land through im-
proved farming practices as a flood control measure is the basis of
the Authority's interest in soil conservation. The impact of the vast
dam-building program on farm lands and farm families necessitated
agricultural planning for readjustment, a problem heightened by the
predominantly rural character of the area. In addition, unified resource
development and conservation were assigned to the Authority as cor-
relative responsibilities. It is important to make clear that while the
7 Testimony before the Senate Judiciary Committee during its lobby investigation
of 1929-1930 revealed (see p. 2892 of the Hearings) that the Farm Bureau was a
consistent enemy of the attempts of Senator Norris to achieve a program of public
ownership as a solution to the Muscle Shoals problem. In 1928, the Farm Bureau
supported President Coolidge's veto of a resolution establishing a Muscle Shoals Cor-
poration, calling the resolution "an abject surrender of what agriculture has been
fighting for at the Shoals in that the production of fertilizers for experimental pur-
poses will not provide agriculture with the highly concentrated complete fertilizers
which are necessary" (p. 2925).
8 The Farm Bureau has been accused of "selling out" the farmers on the Muscle
Shoals issue. See, for example, National Farm Holiday News, organ of the left-wing
Farm Holiday Association, October 30, 1936. However, it appears that the Farm
Bureau abandoned its demands for quantity production of fertilizer at the Shoals
only at the beginning of the Roosevelt administration.
9 Muscle Shoals Hearings, House Committee on Military Affairs (April, 1933), p. 85.
Selznick : T VA and the Grass Boots 89
conception of TVA, held popularly and by President Eoosevelt and
Senator Norris, was broad, the statutory authority provided for fairly
specific activities. The Act did not grant to the TVA any exclusive or
unlimited governmental powers over agricultural development in the
area.
In fact the legal basis which sustains the TVA agricultural program
is contained primarily in Section 5 of the organic Act of 1933, 10 which
provides essentially for the production, development, and introduction
of fertilizer materials. The Board of Directors was authorized,
(a) to contract with commercial producers for the production of such fertilizers or
fertilizer materials as may be needed in the Government's program of develop-
ment and introduction in excess of that produced by Government plants.
This authority has not been utilized ; rather, the second provision,
(b) to arrange with farmers and farm organizations for large-scale practical use of
the new forms of fertilizers under conditions permitting an accurate measure of
the economic return they produce
has represented the core of the program. To implement this responsi-
bility, permissive power was granted
(c) to cooperate with National, State, district, or county experimental stations or
demonstration farms, with farmers, landowners, and associations of farmers or
landowners, for the use of new forms of fertilizers or fertilizer practices during
the initial or experimental period of their introduction, and for promoting the
prevention of soil erosion by the use of fertilizers and otherwise,
an authorization which linked the TVA's program to existing agricul-
tural institutions as well as to the correlative objective of soil erosion
control. Further,
(d) the board in order to improve and cheapen the production of fertilizer is author-
ized to manufacture and sell fixed nitrogen, fertilizer, and fertilizer ingredients
at Muscle Shoals by the employment of existing facilities, by modernizing exist-
ing plants, or by any other process or processes that in its judgment shall appear
wise and profitable to the fixation of atmospheric nitrogen or the cheapening of
the production of fertilizer.
(e) Under the authority of this Act the board may make donations or sales of the
product of the plant or plants operated by it to be fairly and equitably distrib-
uted through the agency of county demonstration agents, agricultural colleges,
or otherwise as the board may direct, for experimentation, education, and intro-
duction of the use of such products in cooperation with practical farmers so as
to obtain information as to the value, effect, and best methods of their use.
These provisions embrace the major framework of the Authority's work
in relation to farmers and agricultural agencies. In addition, provision
10 48 Stat. 58 (May 18, 1933).
90 University of California Publications in Culture and Society
was made for the maintenance of nitrate plant no. 2 in stand-by condi-
tion in the event that nitrogenous fertilizers were not produced, and the
Board was authorized
(h) to establish, maintain, and operate laboratories and experimental plants, and to
undertake experiments for the purpose of enabling the Corporation to furnish
nitrogen products for military purposes, and nitrogen and other fertilizer prod-
ucts for agricultural purposes in the most economical manner and at the highest
standard of efficiency.
Thus, while permitting considerable discretion, the Act itself specified
that the production and distribution of fertilizers should be the basic
contribution of the Authority to agricultural development. Other pro-
visions of the Act relating to agriculture were, however, included. Thus,
Section 4 (1) stated that the Corporation
shall have the power to advise and cooperate in the readjustment of the population
displaced by the construction of dams, the acquisition of reservoir areas, the protec-
tion of watersheds, the acquisition of rights-of-way, and other necessary acquisitions
of land, in order to effectuate the purposes of the Act; and may cooperate with
Federal, State, and local agencies to that end.
While not exclusively an agricultural problem, the acquisition of farm
properties and the need to aid in the relocation and readjustment of
farm families provided further concrete activities for the agricultural
organization within TVA.
Beyond these specific activities, Section 22 of the Act authorized,
through the President, surveys and general plans, as well as "studies,
experiments, or demonstrations" in order "to further aid the proper use,
conservation, and development of the natural resources of the Tennessee
River drainage basin . . . for the general purpose of fostering an orderly
and proper physical, economic and social development " And Section
23 provided that
the President shall, from time to time, as the work provided for in the preceding
section progresses, recommend to Congress such legislation as he deems proper to
carry out the general purposes stated in said section, and for the special purpose of
bringing about in said Tennessee drainage basin and adjoining territory in con-
formity with said general purposes (1) the maximum amount of flood control; (2)
the maximum development of said Tennessee River for navigation purposes; (3) the
maximum generation of electric power consistent with flood control and navigation ;
(4) the proper use of marginal lands; (5) the proper method of reforestation of all
lands in said drainage basin suitable for reforestation; and (6) the economic and
social well-being of the people living in said river basin.
While these general welfare sections of the TVA Act serve to enlarge
the conception of the agency's objectives, they do not in fact grant broad
Selznick : T VA and the Grass Roots 91
powers of action. This has been recognized by the Authority's leader-
ship, resulting in a strong tendency to avoid basing any major activities
upon the authority contained in Sections 22 and 23. The agricultural
program in particular is rooted in the fertilizer provisions of Section 5.
This does not mean, however, that a broad conception of the economic
and social well-being of the rural population has not directed the admin-
istration of the program. On the contrary, the TVA exercises its discre-
tion in the spirit of Sections 22 and 23, as these are variously interpreted.
APPROACHING THE GRASS ROOTS
The first appointee to the TVA Board was its chairman, Arthur E.
Morgan, an engineer of wide reputation and broad social vision. As
engineer and educator, he combined a practical knowledge of water con-
trol problems with an active interest in the general problems of social
reconstruction. His Antioch plan in college education was widely known,
and he had developed a philosophy of responsible public service which
was to play a major role in laying the foundations for the Authority's
personnel system. Morgan's selection by President Roosevelt preceded
the enactment of the TVA statute, and he participated in drafting some
of its provisions.
As chairman and first appointee (though the chairmanship was later
revealed to be of little significance) , Morgan was consulted by the Presi-
dent on the appointment of his codirectors. The President required that
one of the directors be a Southern agriculturist. Morgan approached
representatives of the Extension Service of the Department of Agricul-
ture for suggestions as to a likely candidate from the Tennessee Valley
region. As a result, the President of the University of Tennessee, Dr.
Harcourt A. Morgan, was nominated by the President for the six-year
term. This was a crucial selection, for it materially determined the nature
of the discretion to be exercised by the new agency in its agricultural
program, with attendant consequences for other phases of the agency's
structure and activities. Upon reflection, a few years later, A. E. Morgan
rebuked his own naivete in having felt that elements in the Department
of Agriculture consulted by him would consider professional compe-
tence rather than tangential organizational and political motives in
suggesting a candidate for the TVA Board. However that may be, it
seems clear that the institutional background of H. A. Morgan, particu-
larly as a leader in the land-grant college extension service system,
decisively affected the course of the TVA agricultural program and the
special role of the Agricultural Relations Department among the admin-
92 University of California Publications in Culture and Society
istrative units of the Authority. This is the opinion of informed members
of the T VA staff.
The influence of H. A. Morgan upon the agricultural program was
assured by the early division of functions among the TVA directors."
This followed an initial disaffection within the Board which moved
Directors D. E. Lilienthal and H. A. Morgan to attempt to restrict the
activities of Chairman A. E. Morgan. By this plan, adopted by the
Lilienthal-H. A. Morgan majority in August, 1933, A. E. Morgan was
given responsibility primarily for the general engineering program;
Lilienthal was to be in charge of all matters pertaining to the Author-
ity's power development ; and H. A. Morgan was assigned supervision
of all matters relating to agriculture, agricultural industries, and rural
life, as well as of the chemical engineering program. Thus the Author-
ity's power to exercise discretion in relation to the broad fertilizer and
soil conservation program was concentrated in the hands of the former
president of the University of Tennessee.
Dr. Morgan was the only representative of the Tennessee Valley region
on the TVA Board of Directors. It is fair to say that he represented
the relatively conservative institutional forces in the area, as those were
expressed in the established agencies of state and local government. Dr.
Morgan brought to the TVA a long experience with agricultural work
in the South, together with corresponding professional commitments to
those agencies with which he had been so long associated. It is he who
is the recognized source within the Authority of the grass-roots approach
as a moral enterprise, although public formulations have come primarily
from Lilienthal. As a representative of the area, moreover, Dr. Morgan
brought with him the possibility of mobilizing considerable support for
the TVA from established institutions in the region. It is freely stated
among certain circles within the TVA that the H. A. Morgan-Lilienthal
bloc was a "log-rolling" enterprise, whereby Lilienthal received support
for the electric power program in exchange for his support of the fer-
tilizer program. Whether this was so or not, in terms of conscious be-
havior, is certainly open to question, but it is not unreasonable as an
interpretation of the objective meaning of the bloc. Dr. Morgan's con-
nection with the agricultural extension services, organizations ramified
through every county in the watershed, represented a formidable factor
which might conceivably turn the scale of popular opinion from support
or indifference into antagonism. An extension service official in one of
11 See the "Memorandum on Organization," August 3, 1933, published in Hearings,
Joint Committee to Investigate the Tennessee Valley Authority, 75th Cong., 3d sess.
(August, 1938), pp. 105-107.
Selznick : T VA and the Grass Roots 93
the Valley states stated privately in 1943 that "the TVA investigation
of 1938 would have turned out differently if the extension service had
been alienated. Some witnesses would have testified differently." Another
extension service official said, in 1942, "if the TVA ever tried to go out
on its own, by-passing the extension service, the next appropriation bill
would not come out of committee."
It would not be surprising, therefore, if Lilienthal, recognizing that
the national political implications of TVA were linked above all to its
electric power program, 12 and that this program was facing enormous
practical and constitutional difficulties, might have been willing to
agree to the delegation of discretion in the agricultural program to Dr.
H. A. Morgan. This alignment eventually spelled the doom of such
development programs, supported by A. B. Morgan, as were not ac-
ceptable to the land-grant college group. These included an emphasis on
self-help cooperatives, subsistence homesteads, rural zoning, and broad
regional planning.
The TVA was moving into the agricultural field at a time of transition
in which earlier tendencies of the Department of Agriculture to extend
its activities from education to action were being increased and made
explicit by the new ventures of the New Deal administration. Estab-
lished agencies, notably the land-grant colleges, took heed of possible
threats to their institutional prerogatives, especially those which might
block their nearly exclusive avenue to the farm population through the
county agent organization. This concern was expressed through the
powerful Association of Land-Grant Colleges and Universities, organ-
ized in 1887 for consideration of the problems and interests of the offi-
cials within the state experiment station and extension service network.
In addition, the closely related American Farm Bureau Federation sup-
plemented the activities of the Association from its vantage point as a
mass-pressure group. In 1933, while plans for the initiation of the TVA
agricultural program were being developed, the 47th Convention of the
Association was held, at which "there was much discussion of the work
of the Authority and the relations of the Land-Grant Colleges to it. A
committee was appointed to confer with the Authority 'in regard to
such features of its work as naturally fall within the field of work of
13 "At the heart of the TVA controversy is its power programs. Attacks upon the
Authority are sometimes made in terms of personalities, in terms of its navigation
and flood-control program, and in terms of its fertilizer work or other activities, but
I think the Committee will find that, at bottom, all of the attacks upon the Authority
stem from 'power'." Testimony of David E. Lilienthal, Hearings, Joint Committee to
Investigate the Tennessee Valley Authority, 75th Cong., 3d sess. (August, 1938),
p. 149.
94 University of California Publications in Culture and Society
institution members of this Association/ m3 With H. A. Morgan in
charge of the TVA's agricultural program, the committee could find
little difficulty in arriving at a satisfactory arrangement.
This was not so with respect to other new agencies, however. In 1934,
at the 48th Annual Convention, a report of a "Special Committee on
Duplication of Land-Grant College work by New Federal Agencies"
was heard. This committee pointed to "ample evidence of duplication
(which) invades long-established fields of the land-grant colleges in
extension, in research and possibly in resident teaching." 14 Examples
cited were the Federal Emergency Belief Administration's rural home
demonstration work and the research, demonstration, and extension
work of the Federal Soil Erosion Service. It is significant that prelim-
inary negotiations between the latter agency, then in the Department
of the Interior, and Chairman A. E. Morgan of the TVA, had been ini-
tiated, looking toward a joint program; but the proposal was rejected
by H. A. Morgan, setting the pattern for a selective approach to the
land-grant college system as the primary avenue of cooperation. This
action of H. A. Morgan followed closely the recommendations of the
Special Comnfittee on Duplications that "this Association reaffirm its
willingness to make available the trained and experienced personnel,
the extensive organization and the benefits of the long experience of
the land-grant colleges as the agencies through which Federal-State
cooperation should function in scientific and educational service to in-
dustry, agriculture, and rural life." 15
After two original conferences of representatives of the Valley states
colleges and the TVA in 1933, and a third in July, 1934, a definite pat-
tern emerged within which future relationships would be ordered. The
seven Valley states colleges 18 selected representatives to participate in
joint planning activities, and a correlating committee was created, con-
sisting of three officials, one a joint representative of the Valley states
colleges, one from TVA, and one from the U. S. Department of Agri-
culture. The original committee consisted of Dean Thomas P. Cooper
13 Carleton R. Ball, A Study of the Work of the Land-Grant Colleges in the Ten-
nessee Valley Area in cooperation with the Tennessee Valley Authority, prepared
under auspices of the Coordinating Committee of the USD A, the TVA, and Valley
States Land-Grant Colleges, 1939, p. 12. (No imprint.)
M Proceedings of the 48th Annual Convention of the Association of Land-Grant
Colleges and Universities, Washington, D.C. (Nov. 19-24, 1934), p. 240.
15 IUd., p. 241.
"Alabama Polytechnic Institute, Auburn, Ala.; University of Georgia, Athens,
Ga. ; University of Kentucky, Lexington, Ky. ; Mississippi State College, State Col-
lege, Miss. ; North Carolina State College, Ealeigh, N.C. ; University of Tennessee,
Knoxville, Tenn. ; Virginia Polytechnic Institute, Blacksburg, Va.
Selznick: TVA and the Grass Roots 95
of Kentucky, Director J. C. McAmis of the Department of Agricultural
Relations, TVA, and Dr. C. W. War burton, director of Extension Work
in the U. S. Department of Agriculture. The selection of Dr. Warburton
insured amicable relations, since the entire correlating committee repre-
sented the extension service outlook. Later, another representative of
the Department was to reflect underlying changes in organizational
emphasis. 17
THE MEMORANDUM OF UNDERSTANDING
The correlating committee succeeded in bringing the agricultural activ-
ities of the TVA into line with the established relationships between
the Department of Agriculture and the land-grant colleges. In effect,
this meant the establishment of a major part of the TVA agricultural
program upon a grant-in-aid foundation whereby the TVA supplied
funds for the execution of programs by the states, following the pattern
used by the Department in administering the Morrill, Hatch, and Smith-
Lever Acts. The basic document establishing the three-way relationship
was completed late in 1934 as a memorandum of understanding signed
by the Chairman of TVA, the Secretary of Agriculture, and the presi-
dents of the Valley states agricultural colleges. This agreement was
framed by the correlating committee to provide a "systematic pro-
cedure for a coordinated program of agricultural research, extension,
and land-use planning within the region of the Tennessee Valley Author-
ity." The correlating committee was given formal status, and the employ-
ment of an executive secretary to be jointly financed, was authorized.
Dr. Carleton R. Ball was chosen for this position, with headquarters in
Washington. A system of "state contact officers" was also provided for,
but never implemented.
The core of the agreement was contained in the provision that pro-
posals for joint coordinated activity on the part of any of the partici-
pants would be channeled through the correlating committee. In this
way, the regional implications of proposed projects would be given due
consideration. Beyond this formal objective of the memorandum of
understanding, however, must be considered its implications for the
organizational position of those involved. These implications have been
formulated by informed participants along the following lines :
1. The three-way agreement was a device for the control of TVA or
at least its relevant activities by the Department of Agriculture, which
was at that time dominated by men committed to the land-grant college
17 See below, p. 170.
96 University of California Publications in Culture and Society
system. There was some apprehension that the TYA might embark on
a program of its own, or with some other agency, a fear which found
some support in the attempt of A. E. Morgan to come to an agreement
with Secretary Ickes of the Department of the Interior. While there was
little to be feared from H. A. Morgan and the organization he built
inside the TVA, a formal agreement would bolster the H. A. Morgan
group if pressure should arise from other quarters in TVA for a new
approach. The formal agreement established a precedent and a commit-
ment, and a set of continuing relationships; these would offer a con-
siderable advantage in any controversy which might arise.
2. With the TVA securely within the land-grant college framework,
the agreement could serve as a bulwark against such new agencies as
might wish to work in the Valley outside the framework set by the
colleges. While little recourse would then be possible, still a commitment,
was obtained from the Department of Agriculture as a whole, so that
the correlating committee procedure could be used as a point of entry
and debate, a formal basis from which objections to the new action
agencies as opposed to agencies primarily engaged in research might
be lodged. One TVA staff member said that "the extension service was
scared to death of the trend going toward the action agencies, and they
wanted to stop it in the Valley."
It is noteworthy in connection with this interpretation that in reality
the state organizations, that is, the land-grant colleges, were the strongest
elements among the participants in the agreement. The TVA representa-
tives, as will be elaborated below, actually had the interests of the col-
leges at heart. Having prior commitments to the land-grant college sys-
tem as such, the TVA was not an independent organization bargaining
freely in the light of its own interests and program. The position of
the Department of Agriculture was similarly conditioned since its rep-
resentation was effected through its Office of Cooperative Extension,
which had been unable to retain a status independent of the states. The
Office of Cooperative Extension has been a small organization, relatively
helpless in respect to supervision of the states, and inescapably com-
mitted to the land-grant institutions. It is one of those organizations of
which it has been said that "control runs from the state to the federal
agency rather than the other way." 18 This situation is in part due to,
and emphasized by, the role of the Association of Land-Grant Colleges
and Universities, which tends to "constitute a more coherent body than
18 J. P. Harris, in foreword to V. O. Key, The Administration of Federal Grants to
States (Chicago: Public Administration Service, 1937), p. xii.
Selznick: TV A and the Grass Roots 97
does the Department itself," and has functioned to preserve "the juris-
dictional sphere of the state stations against encroachments by the
research bureaus of the Departments." 19 With such considerations in
mind, it seems fair to say that the memorandum of understanding cannot
be comprehended as an administrative document apart from the context
of informal and tangential goals and relationships within which it was
written. This situation, in which the specific weight of the state organ-
izations turned out to be disproportionately heavy, lends support to
the view that these local institutions found informal representation
within the TVA administrative structure through the H. A. Morgan
group.
The correlating committee has gained some notice in terms of its for-
mal objective of achieving an integrated regional agricultural program. 20
However, there is some doubt that the committee as such has functioned
as a controlling device. The memorandum of understanding is imple-
mented by contracts between the TVA and individual state universities,
including a master contract and specific project agreements. These are
not channeled through the correlating committee, and are not formally
coordinated through any other region-wide institution. "In general,"
said one official intimately connected with the work of the correlating
committee from its inception, "the 3-man committee has not functioned
beyond calling the deans' and directors' conferences. They did have a
meeting about the time the Soil Conservation Service situation got bad,
but this only served to aggravate matters." A parallel interpretation
was given by a member of the committee. It appears that problems of
jurisdiction and liaison have constituted its primary function, although
Valley-wide conferences of deans of colleges and directors of the state
experiment stations and extension services with representatives of the
TVA and the Department of Agriculture have been held semiannually.
Minutes of the conferences indicate that many substantive and adminis-
trative problems of mutual concern have been discussed, and this has
doubtless contributed to a heightened regional self -consciousness as well
as to a regional exchange of information and ideas. An examination of
the record of these conferences indicates that the TVA representatives
have not exercised the strong leadership which their special regional
responsibilities might seem to imply. The TVA representatives have
19 Key, op. tit., pp. 179, 201. See also G. A. Works and Barton Morgan, The Land
Grant Colleges, Staff Study No. 10, Advisory Committee on Education (Washing-
ton: U. S. Govt. Printing Office, 1939), p. 107.
20 See John M. Gaus and Leon O. Wolcott, Public Administration and the U. S.
Department of Agriculture (Chicago: Public Administration Service, 1940), p. 371.
98 University of California Publications in Culture and Society
consciously held back, on the assumption that a Valley program must
come from the states, on their own initiative. One TVA staff member
(not, however, connected with the agricultural unit) contended in 1943
that "a truly regional agricultural program has not been achieved.
Rather, the methods have followed the general pattern of federal-state
relations in the administration of federal grants-in-aid. Such a pattern
would allow for duplication among the activities of the states." In the
absence of notable accomplishments in respect to the formal objectives
of the three-way agreement, it would seem that the informal implica-
tions, noted above, gain increased significance.
THE DECISION FOR PHOSPHATES
As we have seen, the first major act of discretion exercised by the
Authority in relation to its agricultural responsibilities was the estab-
lishment of cooperative relationships with the land-grant colleges in the
Valley. Concomitantly, it was necessary to decide whether the fertilizer
program of the Authority would revolve around the production of
nitrates, as appeared to be implied in the Act, but was clearly not man-
datory, or some other fertilizer product. On this, the position of the
land-grant colleges and the American Farm Bureau Federation had
already been made clear : the combination of lime, phosphate, and le-
gumes would break the cycle of soil exhaustion brought on by the use
of commercial nitrates under soil-depleting crops. These crops (includ-
ing cotton, corn, and tobacco) are clean tilled, so that emphasis upon
them in the agricultural economy adds to the problem of soil erosion,
which in turn means increased flood hazards. It was not surprising,
therefore, that, when the TVA approached the land-grant colleges for
advice, the result was an exclusive emphasis on the production of phos-
phatic fertilizers. The contribution of TVA was to be the development
and production of concentrated phosphates in order to reduce the burden
of shipping costs, together with the distribution of large quantities of
the new materials for practical testing and demonstration of their use-
fulness. The latter phase constitutes the agricultural as distinct from
the chemical engineering program of the Authority.
The emphasis on phosphates has received wide support in agricultural
circles, and surprisingly little criticism, in view of the striking abandon-
ment of work on nitrates. 21 Criticism, however, was voiced to the Joint
21 The TVA was directed in its Act to maintain nitrate plant no. 2 in stand-by
condition. But the phosphate emphasis apparently strongly modified even the national
defense implications of the nitrate program, for TVA's chemical contribution to
World War II was primarily elemental phosphorus.
Selznick : T VA and the Grass Roots 99
Committee Investigating the TVA by the Board of Trade in Sheffield,
Alabama, that the immediate needs of poor farmers can be met only by
cheap nitrogenous fertilizers, since such farmers are unable to afford a
program of crop rotation and nitrogen fixing through the use of legumes.
Such a view finds support in the statements of some Farm Security
Administration supervisors that it is impossible, in view of the pressing
needs of their clients for a cash crop, to avoid making loans for the
purchase of commercial nitrates. The possibility has also been suggested
that the phosphate program may have been a device for turning aside
government competition from nitrates, as the chief interest of the ferti-
lizer industry. No evidence could be found to support this charge. More-
over, the fertilizer industry is less than enthusiastic about even the
present TVA program, and there is evidence that pressure was brought
to bear upon the Agricultural Adjustment Administration, in 1939
and 1940, to suspend the distribution of phosphates as grants-in-aid to
farmers. Whatever the reasons, soil conservation or otherwise, the turn
to phosphates was made, in line with the program of the land-grant
colleges. 22
EXTENT OF THE PROGRAM
In the light of the long Muscle Shoals controversy, which involved per-
sistent agitation for the production of cheap fertilizers, it should be
pointed out that the TVA has not significantly affected the over-all
output of the fertilizer industry. The Authority restricted itself to an
experimental program, although the initial interest in the fertilizer pos-
sibilities of Muscle Shoals centered around large-scale production of
nitrates. Indeed it was the fear that the offers of Henry Ford and the
American Cyanamide Co. did not include a bona fide interest in the
quantity production of nitrates which was the source of considerable
22 The TVA chemical engineers believed that the manufacture of nitrates was un-
feasible with the plant taken over by the Authority at Muscle Shoals. This may have
been the immediately decisive factor in reaching the decision to manufacture phos-
phates rather than nitrates. See Hearings, Joint Committee to Investigate the Ten-
nessee Valley Authority, p. 1289. However, the decision is usually defended not on
those grounds but on the theory that regional needs were best served by a phosphate
program. It would be difficult to understand why the Authority did not come to Con-
gress with a program for such alterations in the fertilizer plant as would make
nitrate production technically feasible (since that was widely thought to be the
responsibility of the Authority) were it not that the decision was taken on nonengi-
neering grounds. In this connection, H. A. Morgan reported that in 1933 he held "a
series of conferences with officials in the Department of Agriculture, representatives
of the experiment stations and extension agencies of the land-grant colleges, and
officers of the National Association of Land-Grant Colleges. They advised the Author-
ity to produce a limited amount of specific types of concentrated phosphate fertilizer
with which to carry on tests and experiments under controlled conditions." Hearings,
Joint Committee to Investigate the TVA, op. oit., p. 122. (Emphasis supplied.)
100 University of California Publications in Culture and Society
opposition to them. It was believed that the major interest of these cor-
porations was in the electric power possibilities of the installations.
Ironically, concentration on the electric power issue appears to have
resulted in limited fertilizer production at the Shoals even under public
ownership. Whatever the merits of the issue, it seems somewhat anti-
climactic that, after the long debate as recorded in voluminous Congres-
sional hearings, the issue of cheap fertilizer should have been resolved
TABLE 1
DISTRIBUTION OF TVA FERTILIZERS TO JANUARY 1, 1943
For Experiment Stations To
Ten different materials used in preliminary investigations 5 , 570
For Test-Demonstration Farms
Triple superphosphate 149,810
Calcium metaphosphate 33,908
Fused rock phosphate 1 , 218
Total 184,936
Calcium silicate slag, a liming-material by-product 3,008
For Agricultural Adjustment Administration
Triple superphosphate 240, 198
Calcium silicate slag 341,201
SOURCE: Test-Demonstration Information Book, published by the Land-Grant Colleges and
Universities of the Tennessee Valley States and the Tennessee Valley Authority (March, 1943),
p. 109.
by an act of administrative discretion. It is doubtful that it could have
been so resolved without the consent of powerful political interests in-
volved in the controversy. The curious elimination of the strong pressure
for a large nitrate program at the Shoals may be explained in part,
however, by the fact that electric power became the crucial element in
the struggle over TVA as a whole, so that little attention was given to
the fertilizer program by the managers of the liberal-progressive forces.
Table 1 summarizes the amount of materials distributed by the Au-
thority to January 1, 1943. Later, in connection with the resumption
of nitrogen manufacture for war purposes, some nitrogenous fertilizer
was produced and distributed. However, during the first decade of its
operation, the TVA fertilizer program was almost exclusively devoted
to phosphates. The 1939 Census of Manufactures 23 reported that the
23 Vol. II, Pt. I, p. 801.
Selznick : T VA and the Grass Roots 101
fertilizer industry had produced that year 5,088,468 tons of mixed fer-
tilizer and 4,152,642 tons of superphosphate. TVA production, while
doubtless of real significance over the long run, through the introduction
of new concentrated forms, has affected neither this total output nor
the general price of fertilizer.
TABLE 2
SUMMARY OF PAYMENTS UNDER COOPERATIVE CONTRACTS WITH STATE AND
LOCAL INSTITUTIONS, FISCAL YEAR, 1942
Department Amount
Agricultural relations $ 746,048
Commerce 78,555
Reservoir-property management 131 , 485
Chief engineer 612
Health and safety 71,580
Personnel 2,200
Regional studies 5, 700
$1,036,180
SOURCE : TVA Budget Office.
TABLE 3
REIMBURSEMENTS MADE TO LAND-GRANT COLLEGES UNDER COOPERATIVE
AGREEMENTS FOR PERIOD JULY 1, 1935, TO JANUARY 1, 1943
Total Master agric.
Institution reimbursed contract
Alabama Polytechnic Institute $ 699 , 048 . 76 $ 504 , 046 . 87
Georgia Universities 402,137.16 312,226.96
University of Kentucky 231,255.99 199,776.93
Mississippi State College 231,675.52 178,289.32
North Carolina State College 434,498.59 402,041.41
University of Tennessee 2,283,116.92 1,874,146.62
Virginia Polytechnic Institute 322,849.07 317,931 . 15
$4,604,582.01 $3,787,459.26
SOUEOB : TVA Auditing Section.
The predominance of the agricultural program among those exempli-
fying the grass-roots approach within TVA is indicated in table 2, which
lists reimbursements by TVA under cooperative contracts with state
and local institutions in 1942. In table 3, total reimbursements to the
colleges and those made under the master agricultural contracts are
reported, covering 1935 to 1943. These are useful because the scope
102 University of California Publications in Culture and Society
of the TVA agricultural program cannot be gauged simply from a
study of the bulk of materials distributed. In its educational effort, the
employment of personnel has been of considerable importance, consti-
tuting the major factor in reimbursements by TVA to the colleges.
A good indication of the extent of the TVA agricultural program, as
it operates through the state institutions, may be derived from the num-
TABLE 4
PERSONNEL ON COOPERATIVE TVA-LAND-GRANT COLLEGE AGRICULTURAL
PROGRAM, SALARIES REIMBURSED BY TVA IN 1942
Fertilizer Operations Test Demonstrations
Contractual: members of college staffs
Inside Valley states 171
Outside Valley states 30
Direct: Members of TVA staff 11
Fertilizer Operations Distribution
Direct: members of TVA staff 6
Fertilizer Operations Program Exposition
Direct: members of TVA staff 5
Preliminary Investigations (Experiment Stations)
Contractual: members of college staffs
Inside Valley states 35
Soil Inventories and Land-Use Studies
Contractual: members of college staffs
Inside Valley states 34
Contractual: Bureau of Plant Industry 3
Direct: members of TVA staff 7
Readjustment Program
Contractual: members of college staffs
At reservoir-affected areas 63
Direct: members of TVA staff 5
SOUEOE: Budget of Agricultural Relations Department, TVA. Figures on TVA include both
supervisory and clerical personnel.
ber of officials of the land-grant colleges assigned to the TVA program,
with salaries ultimately paid by the Authority through a reimbursement
procedure. These figures are reported in table 4.
Personnel at the professional level in the Agricultural Relations
Department of TVA is limited to about thirty-one individuals. This
parallels the pattern of a small headquarters staff evidenced in the rela-
tion between the Washington Office of Experiment Stations and Exten-
sion Service and the land-grant colleges. The program as a whole,
however, is carried on by a network of TVA subsidized personnel sta-
tioned in every county within the watershed. Most of this personnel is
Selznick: TVA and the Grass Roots 103
concentrated in the test-demonstration and readjustment programs, rep-
resented primarily by assistant county agents working out of the local
county agents' offices.
TABLE 5
TVA MATERIALS INVESTIGATED BY EXPERIMENT STATIONS IN THE
TENNESSEE VALLEY TO 1943
Tons
Triple superphosphate 3, 166
Calcium metaphosphate 1 , 760
Calcium silicate slag 69
Fused phosphate 460
Dicalcium phosphate 32
Phosphoric acid (liquid) 23
Raw rock phosphate 15
Monocalcium phosphate 13
Potassium metaphosphate 32
Potassium-calcium metaphosphate 3
SOURCE: Test-Demonstration Information Book, p. 85.
TABLE 6
TEST DEMONSTRATION FARMS, AS OF JANUARY 1, 1943
Unit
Area
In Valley states
farms
farms
Total
Inside Valley
6 520
12 974
19 494
Outside Valley
3,194
3 194
In non- Valley states. .
4 545
110
4 655
14,545 13,084 27,343
Total area of 27,343 test-demonstration farms active January 1, 1943: 4,005,854
acres.
Total area of 41,951 test-demonstration farms established to January 1, 1943:
6,232,207 acres.
SOUBCB : Test-Demonstration Information Book, p. 107.
From the funds and personnel thus expended has been elaborated a
program of plant food testing, farm management education, and popu-
lation readjustment, to name only the major phases of the work. 24 The
experiment stations of the land-grant colleges have made numerous tests
of concentrated phosphatic fertilizers manufactured by the Authority.
In 1943, it was reported that a total of over 5,500 tons of experimental
34 This study does not purport to present an adequate description or appraisal of
the substantive program. These data are introduced only to provide reference points
for further discussion of administrative relationships, machinery, and problems.
104 University of California Publications in Culture and Society
fertilizer materials had been subjected to such investigation. This figure
is broken down in table 5. An indication of the extent to which the pro-
gram reaches the farm population is presented in table 6, from which it
will be noted that 27,343 farms in twenty-eight states were involved in
TABLE 7
NUMBER OF FAMILIES REMOVED FROM TVA RESERVOIR AREAS TO MARCH 1, 1943
Property
Area (by dam)
Total
owners
Tenants
Norris
2,899
1,708
1,191
Wheeler
842
61
781
Pickwick
506
126
380
Guntersville
1,182
152
1,030
Chicamauga
903
263
640
Hiwassee
261
76
185
Watts Bar
832
196
636
Cherokee
875
376
499
Chatuge
278
166
112
Nottely
91
26
65
Appalachia
22
4
18
OcoeeNo. 3
0*
Douglas
525
201
324
Fort Loudoun
273
91
182
Kentucky
l,327 b
511
816
Fontana
482
14
468
South Holston d
71
37
34
Watauga
43
28
15
11,412 4,036 7,376
SOURCE : Reservoir Property Management Department, TVA.
Ocoee No. 3 Reservoir presented no family removal problem. The Authority acquired the land from
Tennessee Electric Power Co.; in August, 1940, it was transferred to U. S. Forest Service, TVA retaining
the right to flood it for Reservoir purposes.
b Kentucky Reservoir 51 per cent evacuated.
Fontana Reservoir 40.1 per cent evacuated.
d Cessation of construction ordered by War Production Board.
the fertilizer distribution activities. Table 7 reports the number of fami-
lies removed from reservoir areas, as an index of the extent of the read-
justment activities.
THE TVA MACHINERY
The major administrative arm of the Authority in the field of agricul-
ture is its Agricultural Relations Department. It is this organization
which carries on the cooperative fertilizer program, particularly those
aspects which directly affect the agricultural population. As we have
seen (table 2), its contractual relations with the land-grant colleges ac-
Selznick : T VA and the Grass Roots 105
count for the major portion of funds used within the framework of the
grass-roots approach.
Before discussing the character of the Agricultural Relations Depart-
ment, some attention should be paid to its place within the general,
formal TVA structure. Chart 1 (p. 106) reproduces the official organi-
zational chart of TVA, and chart 2 (p. 107) abstracts those units which
are delegated responsibilities relevant to the agricultral program. In-
dicated upon the latter will be noted :
Chief Conservation Engineer. This post was created in 1937, con-
comitantly with the general reorganization of the Authority, and sym-
bolizes the unified approach to "water control on the land." Placed in
charge of three subject-matter departments, this official was assigned
the responsibility of coordinating "the Authority's program of national
defense, laboratory research, experimental production, and test of fer-
tilizer and fertilizer ingredients, agricultural demonstrations and for-
estry conservation studies and activities, as carried on by the Chemical
Engineering, Agricultural Relations, and Forestry Relations Depart-
ments. . . ," 25 C. Niel Bass, a former official of the city of Knoxville, was
appointed to this post, and has been its only incumbent. He was origi-
nally brought into the Authority as an administrative assistant to H. A.
Morgan. In an appraisal of the working relationships, the position of
Chief Conservation Engineer does not have the de facto significance
which its formal status would seem to imply. Little effective control over
the Agricultural Relations Department is exercised, if any was intended.
On major issues, Bass tends to follow the lead of the agriculturists, as
that is formulated by Director H. A. Morgan and J. C. McAmis, head
of Agricultural Relations. It is believed in some quarters that Bass's ap-
pointment was in effect a means of consolidating the dominant position
of the agriculturists in relation to kindred subject-matter departments.
This interpretation gains strength from the history of conflict involving
the early leadership of the Authority's forestry program, which was
linked to the A. E. Morgan group. With the departure of E. C. M.
Richards, Chief Forester, as part of the turnover of leading personnel
following the ouster of A. E. Morgan, it was necessary to reintegrate
the remaining members of the forestry division. The coupling of the
reorganized Forestry Relations Department to the leading supporters
of H. A. Morgan may reasonably be interpreted as a device chosen for
this Gleichschaltung.-
25 Admin. Bulletin No. 23, Office of the Gen. Manager, TVA, August 16, 1937
(mim.).
28 See below, pp. 147 ff.
O
11
-2 2
m S
13 53
M
s a
"K
n
t
If?
OQJ o M
i ^| ^
L_
ti)
s
^ Q. O
III 1
*
VM.fi O
| lls ll i] 1
& n^ ... 5j o^ & ^3 Q o ^ o
.s
O <S +2 H
A
I -|
0^ S^^ tJ.r2 t^>^ "C
s
H
M 5 H " 0^5?^ 0^ S
.s
O
^ ^*^ ^ ^**s ^ /**^ ^-i Q
^
M
cd ^p{ ^* O ^
s
o
^ Q PH
1
3
h
to CM 3*^ H
?
*O '^a *O 'C -2 O .2
1
| ?! 1| Ifl l|
c3
^3pH *Jp -gjl g -gO
oo C^o> ^J^^ P- 1
,
8 SI SI'S SI
* M _^_
-CO'
a l
s-3
^ w
^
^n'bJO
Sin
If
E
a
w
1
S *1 <
o> be
00
SQ 0^
s
J
^
f
a
S
-J-
T
n
Management
Service
Council
1 Personnel
M Department
H Finance
Department
a!
fi
^Land
Acquisition
Department
1
T
<*>>*&
P S
o
H'
o ce
g"3
O.g
I
108 University of California Publications in Culture and Society
Department of Chemical Engineering. This is the Authority's tech-
nical organization for research in and production of fertilizer materials.
It is in direct charge of the physical plant at Muscle Shoals. A Phosphate
Division conducts mining operations in the TVA phosphate properties
in middle Tennessee; a Chemical Operations Division constructs and
operates the commercial-size plants for fertilizer (and munitions) in-
gredients; a Chemical Engineering and Design Division contributes
designs for the Operations Division and provides miscellaneous engi-
neering services ; and a Chemical Research and Development Division
operates a laboratory-to-pilot-plant research program. This work is
supplemented by cooperation with the University of Tennessee in the
maintenance of a Chemical Engineering Research Laboratory and the
Cooperative Chemical Research Unit at the University's Experiment
Station. The basically technical character of the functions of the Chemi-
cal Engineering Department was underlined in 1941, when the Com-
merce Department was made responsible for coordinating the industrial
research activities of the Authority and for their articulation with the
research activities of other agencies. 27
Department of Forestry Relations. Chiefly through its Watershed
Protection Division, this organization is related to the agricultural pro-
gram by virtue of its responsibility to cooperate with other local, state,
and federal agencies for erosion control, reforestation, farm forestry,
and forest protection. Cooperation with the state extension service or-
ganizations has been instituted, though not in the same manner or with
the same consequences as with the Agricultural Relations Department. 28
Technical supervision for camps operated by the Civilian Conservation
Corps has been provided, and nurseries are operated to produce tree
seedlings and explore and test special tree crop species. This Department
also includes a Forest Resources Division, which conducts studies for
developing the forest wealth of the Valley. A Biological Readjustment
Division furthers the wild-life resources of the region.
Commerce Department. This was established in its present form in
1939, by consolidation of an earlier Commerce Department, having
more restricted functions, with the former Department of Agricul-
tural Industries. Continued agricultural interest was manifested in
the responsibility which the Board delegated to it to conduct research
in the development of new agricultural equipment, income-producing
rural electrical equipment, and the development and stimulation of
27 Admin. Code No. 41-0, Office of the Gen. Manager, TVA, Oct. 7, 1941 (mim.).
28 See below, pp. 147 ff .
Selznick : T VA and the Grass Roots 109
rural electrification education. In line with the grass-roots policy, the
Department was enjoined to "carry on its activities in such a manner
as to stimulate the efforts of other agencies in the region on a permanent
basis, either through cooperative formal or informal understandings or
agreements, and with due regard for the needs and facilities of the
cooperating agencies." 29
Reservoir Property Management Department. Responsible for the
administration of TVA-owned lands, this organization maintains rela-
tions with the state extension services, which especially in the organi-
zation of the local Land-Use Associations, through which rents are
collected and conservation practices advanced aid in dealing with those
farmers who rent Authority land for agricultural purposes.
Formal Structure of the Agricultural Relations Department. Oper-
ating on a budget of well over one million dollars yearly, this unit within
TVA is the foremost representative of the grass-roots approach, both in
its own vigorous espousal of the doctrine and in view of its central role
in the TVA's relation to the farm population. As we have already noted,
the headquarters staff is small, and no TVA field offices are maintained,
the entire operating load being carried by the land-grant colleges of the
Valley area. The bulk of the department's available funds is disbursed
to the colleges virtually on a grant-in-aid basis. Chart 3 (p. 110) repro-
duces the official organization chart of the department.
The Test-Demonstration and Reservoir Adjustment Divisions bulk
largest in the agricultural program. The former is responsible for super-
vision of the arrangements with the land-grant colleges under which the
Authority's fertilizer products are supplied to farmers for conducting
unit and area test-demonstrations on representative farms. The TVA
staff for this work numbers only four professional and seven clerical
personnel, while a total of 201 TVA-subsidized supervisors, specialists,
and assistant county agents are employed by the colleges. 30 The Reservoir
Adjustment Division is responsible for providing assistance and advice
"with regard to problems of land use, and to relationships involving or
affecting agricultural institutions and groups where such problems are
associated with land acquisition, the readjustment of population in
reservoir-affected rural communities and the management of TVA-
controlled lands." 31 Four supervisory officials are on the Agricultural
29 Admin. Bulletin No. 25-1, Office of the Gen. Manager, TVA, July 24, 1939 (mim.).
30 Of these, 171 were in the seven Valley states, and 30 in twenty non-Valley states.
31 Admin. Bulletin No. 43-1, Dec. 10, 1940, Office of the General Manager, TVA
(mim.).
:f
gP
I
go
li
el
111
&s
fl
W
S
il
9U
r
i
I
a
Selznick : T VA and the Grass Roots 111
Relations Staff at Knoxville, while sixty-three are employed through the
land-grant colleges to work in reservoir-affected areas.
The Program Exposition Unit is a small group providing educational
materials for farm groups and the general public. The Preliminary In-
vestigation and Survey Division deals primarily with the experiment sta-
tions in relation to soil surveys and controlled tests of fertilizer products.
The Fertilizer Requests Division is a service organization, facilitating
TABLE 8
ALLOTMENT OF FUNDS, BY PROGRAM AND METHOD OF DISBURSEMENT, WITHIN
TVA AGRICULTURAL RELATIONS DEPARTMENT (1943-1944)
Program
Test demonstration
Contractual
$ 507,000
Direct
$ 29,000
Readjustment and relocation
372 000 a
25 000 a
Fertilizer requests
19,000
Program exposition
25 000
Soil and fertilizer investigations
90 000
28 000
Soil inventory
.. . 87,000
18 000
Cooperatives
25 000
General and administrative. . .
79.000
Total budget: $1,304,000 $1 ' 056 ' $248 ' C
SOURCE: Proposed departmental budget, TVA, Department of Agricultural Relations, 1943-
1944.
a Approximate.
the routing, dispatching, and accounting control of fertilizer material
distributed in connection with the test-administration program. A small
Rural Cooperatives Research Division 32 provides assistance in relation
to those programs of the Authority which involve the development of
cooperatives, in particular the county soil associations and the rural
electric cooperatives. Relative importance of divisions is shown in table 8.
Character and Role of the Agricultural Relations Department. Like
Harcourt A. Morgan on the TVA Board of Directors, the Agricultural
Relations Department functions as an institutional representative of
the Tennessee Valley region within the Authority. In particular, it func-
tions as a representative of a special local leadership : the land-grant
colleges and the established farm leadership. An interpretation which
ignores this relation cannot help being weak and unrealistic; on the other
hand, to make use of this principle is to be in line with those within the
TVA who are well informed.
32 See below, p. 230.
112 University of California Publications in Culture and Society
The department is a reflection of the views and personality of Director
H. A. Morgan. This is assured by the close personal relation of the
department head to Dr. Morgan and by the institutional origins of the
other leading personnel who administer the agricultural program. Not
unnaturally, the former president of the University of Tennessee chose
a member of the university farm management staff, J. C. McAmis, to be
his second-in-command in developing the Authority's agricultural ac-
tivities. McAmis joined the TVA staff in 1933 and since then has been
in continuous charge of the Department of Agricultural Relations and
its earlier homologues. Like Dr. Morgan, he is a vigorous advocate of
the land-grant college system, and a strong defender of the South against
encroachments by those inclined to come with what appears to him to
be more fervor than understanding. His loyalty, and that of others on
the staff, to "Dr. H. A." is often expressed and emotionally buttressed
by the evangelical halo which surrounds Dr. Morgan's attitude to the
problem of "man and phosphate."
The leadership of the department is concentrated in the hands of
four or five men whose professional roots lie primarily in the Tennessee
extension service. Thus the major phases of the TVA fertilizer program,
so far as it is brought directly to the farmer, are administered by a group
of men whose attitudes and social commitments have been framed by
forces within the Valley area. It is a group which is strongly conscious
of the need to preserve the integrity of the extension service organiza-
tions, on the state level, in all relations with the TVA.
The institutional origins of this leading group imply a set of basic
attitudes of great significance in the administration of governmental
programs. These attitudes, among others, relate to the Negro population
in the South and to farm tenancy. There is evidence that the typical
position of the TVA agriculturist on the former is one of white superi-
ority. This is supported by references to "good and bad niggers," an-
tagonism toward those who "meddle" in such issues, as well as by the
typical refusal to accord to Negroes such symbols of courtesy and mutual
respect as the title "Mr." On the problem of land tenancy, the prevailing
attitude is one of paternalism, with the assumption that landlords should
"take care of" their tenants, who are generally deemed to be satisfied
with their lot. The regional pattern of relative unconcern for the special
position of the Negro population is carried out in the exercise of discre-
tion in utilizing local institutional resources. The Negro agricultural
colleges have had no place in the TVA fertilizer program, and the op-
portunity to increase the number of Negroes on the staff of the extension
Selznick: TVA and the Grass Roots 113
program in the field has not been utilized. 33 Negro farmers have been
involved in the program through the regular county-agent system work-
ing out of the white universities, but no special effort has been made by
the TVA agriculturists to strengthen the Negro colleges. Traditional
attitudes toward tenancy may be related to the hostility of the depart-
ment to the tenant-oriented Farm Security Administration. 34
These responses of the TVA agriculturists are representative of those
prevailing within their area of operation. Consequently, in one view,
they may be considered to legitimately implement the grass-roots method.
Yet it may be suggested that the Authority might have picked a leader-
ship with a more forward-looking approach to pressing rural problems,
and cautiously attempt to induce the local institutions to show greater
concern for the relatively dispossessed elements. It might be suggested
that the Authority need not have chosen men who would simply reflect
prevailing institutional attitudes ; a local leadership need not imply a
leadership totally acquiescent in established inequalities and their sup-
porting codes. Indeed, in the early days of the Authority, some effort
was made to make official contact with rural leadership tending in a dif-
ferent direction. But this did not come on the initiative of the agricul-
turists within the TVA and came to nothing.
If, however, the TVA grass-roots policy, in its agricultural phase, is
viewed in the light of its organizational function, the consistency and
reasonableness of the character of the Agricultural Kelations Depart-
ment become clear. It was suggested in the preceding chapter that the
meaning of the grass-roots doctrine might be interpreted significantly
in terms of the need of the TVA to adjust, to make its peace, with at
least a major segment of the local institutions. As the TVA is an imposed
33 Of the land-grant colleges, Barton and Morgan, op. cit., p. 101, report that ". . .
after allowance has been made for such factors as the utilization of the services of
white county agents in behalf of Negroes, interracial meetings, and the segregation
in research reports of the data for the races, and the more recent attempts in the
work of the experiment stations to give attention to the rural problems peculiar to
the Negro, there is still evidence that in research and extension there has been dis-
crimination against the Negro. This appears to be more marked in extension than
in research. Experience under the several land-grant acts points toward the need of a
Federal safeguard if discrimination on racial grounds is to be prevented." In 1936,
J. Max Bond, then responsible for Negro personnel work in the TVA, called attention
to the opportunity for bettering the relative position of Negroes in the extension
service, without favorable response. The issue was also raised by the National Asso-
ciation for the Advancement of Colored Peoples at the time of the Congressional
Investigation of TVA in 1938. See Hearings, Joint Committee to Investigate the
TVA (op. cit., in n. 11 above), pp. 2383 ff. That the department is conscious of
unfavorable criticism is evident from the suggestion by a TVA agriculturist that
work of a white assistant county agent in a small Negro community be given a special
publicity "build-up."
34 See below, pp. 164 ff .
114 University of California Publications in Culture and Society
agency born in the midst of and faced with a long struggle against
powerful interests in the electric power industry, cognizant of the poten-
tial sources of hostility in sectional self -consciousness and states-rights
sentiment, and committed on a national scale to the predominant impor-
tance of its electric power program the conditions of its existence easily
support the interpretation that the fertilizer program was turned over
to the existing leadership in the area in exchange for at least tacit sup-
port of the power program. Whatever the motives of the leading partici-
pants, this appears to be the objective import of the organizational
relationships which were established.
The self -consciousness of the Agricultural Relations Department in
respect to its social origins is remarkable, and may be referred to a defen-
sive psychology generated by being part of a larger organization with
quite different roots. The department chief once said, significantly,
" , myself, and are products of a system you can take
your coat off when you come into TVA, but you can't take your hide off."
Faced with the need to defend the region and its institutions against
alien elements, it is not surprising that the TVA agriculturists should
have developed a special loyalty within the organization. This defense
is especially important because social origin does not appear to be ade-
quate as an explanation for the institutional role of this group inside
TVA. Other farm agencies, such as the Agricultural Adjustment Admin-
istration, Farm Security Administration, and Soil Conservation Service
have recruited professional personnel from the local extension service
staffs, without, so far as is known, developing the same self -consciousness.
This may be a matter of training in service. It is probable that H. A.
Morgan and J. C. McAmis, having uncontested control over the agri-
cultural program, so defined the situation for new personnel that the
question of defending the integrity of local institutions became para-
mount in the minds of the basic agricultural staff. In other agencies,
such as Soil Conservation Service, a determined effort is made to reorient
personnel so that the viewpoint of the leadership may be suffused
through the ranks. Some SCS men, recruited from extension service,
reported that they "really learned about soil conservation since joining
the SCS." An analogous statement, with implied deprecation of the ex-
tension service, would not find favor in the eyes of the TVA agricul-
tural leaders.
The position of the Agricultural Relations Department is reinforced
by the theory of TVA central management which values decentralization
within the Authority. The General Manager's Office is committed in
theory to permit subject-matter departments a wide measure of discre-
Selznick: TVA and the Grass Roots 115
tion in the administration of their programs. In one sense, this only of-
ficially validates the strong precedents, established in the early days of
the Authority, for a "hands-off" attitude toward the agricultural pro-
gram. Doubtless, however, the fact that the department has virtually
its own representative, H. A. Morgan, on the three-man Board of Direc-
tors must underlie all other considerations. Some elements in and around
central management have nevertheless attempted to bridle the depart-
ment with tentative and somewhat hesitant attempts to institute controls
from above. "We tried," said one man close to central management,
"putting someone above him, below him, and alongside him." The "him"
refers to the Director of Agricultural Relations, and the other persons
to, respectively, the Chief Conservation Engineer, 85 an assistant director,
and a consultant. None of these efforts was effective, however. It is sig-
nificant that in at least one case the aggressive and evangelical person-
ality of the department chief was able to shape the viewpoint of a young
administrative assistant, who is supposed to have been sent in by central
management. As a consequence, when this man was later transferred
to a strategic post in central management, it was possible for the depart-
ment to have an ideological representative in the General Manager's
office, writing memoranda on controversial matters.
On the whole, however, the position of central management is not un-
ambiguous. It is by no means clear that the General Manager himself,
keenly aware of the problems involved, is (1943) behind or supports all
attempts by elements in his office to strengthen formal control over
the Agricultural Relations Department. Thus, when in 1943 the TVA
Budget Office sought to bring about tighter control over contractual rela-
tions with the land-grant colleges, and to strengthen the role of central
management in the administration of those contracts, there was some
doubt as to the seriousness of the top leadership itself. This was not lost
to the Agricultural Relations Department. In commenting on the report
of the Budget Office, one of the leading agriculturists suggested :
It seems to me that the important thing for us to do first is to find out to what extent
the General Manager's office considers this report of importance. How much attention
should we give it ? It may be that the General Manager's office knows certain things
should be done about which we cannot be informed.
It appears that the General Manager and his closest associates are more
inclined to deal in personal and informal terms with department heads
than to rely on formal controls.
The Agricultural Relations Department, however, is not unwilling to
use formal mechanisms to further its special interests. Thus, in 1943, the
35 But see above, p. 105.
116 University of California Publications in Culture and Society
department tried to have central management promulgate an official ad-
ministrative code for the "coordination of the activities and relationships
pertaining to agriculture" which in general would give to the Agricul-
tural Relations Department the power to coordinate all regional develop-
ment programs undertaken by the Authority and affecting agriculture
in any way. This would include programs carried on by the TVA Com-
merce Department such as the processing of food and fiber products. It
would channel through the department all contact by the Authority with
farm people. Its significance is that the strong emphasis of the proposed
code on dealing with the local institutions would assure the position of
land-grant colleges. Relationships instituted voluntarily and enthusias-
tically by the agriculturists in TVA would become virtually an enforce-
able administrative rule for other TVA departments. 80
Rumblings of a more serious nature, sufficient to shake the equanimity
of the agriculturists, have occurred, however. In September, 1943, at a
regular joint conference of the Board of Directors and representatives
of management, Director Lilienthal is reported to have expressed con-
cern over the economic and social phases of agricultural development in
the Valley. From the point of departure of the report of a Congressional
committee on industrialized farming in California, the problem of pos-
sible changes in the pattern of farm ownership in the Tennessee Valley,
and especially the place of the family farm, was discussed. It was sug-
gested that the present TVA agricultural program might be inadequate
in face of such possible changes. The background of the TVA agricul-
turists has inhibited serious consideration of such problems, so that it
is safe to say that the department was caught off guard by the remarks of
the then TVA Chairman. Lilienthal, it appears, considered it a function
of TVA management to provide leadership for the region. But the agri-
culturists inside the organization incline to the view that they "make no
effort to lead development in any particular direction" and that, indeed,
they have no special program distinguishable from that of the land-grant
colleges in the field of agriculture. From trends observable in 1942-1943,
however, it is probable that the department will find it expedient to
undertake a broader measure of initiative, particularly in the develop-
ment of farmers' cooperatives. 87
36 See p. 151.
37 As TVA phosphate production increased, the problem of distribution through
new mechanisms, beyond the test-demonstration and other government programs, was
posed. The result was an important victory for the supporters of farmer coopera-
tives. In 1944, the TVA publicly affirmed its support of cooperatives as having an
important role in assuring an adequate national supply of fertilizers. See the state-
ment of the Board of Directors, "Mineral Fertilizers and the Nation's Security," De-
cember 1, 1944.
CHAPTEKIV
TVA AND THE FARM LEADERSHIP
(Continued)
In analyzing the relations between TVA and the extension service, the significant
question for you to ask is whether the tail has not begun to wag the dog.
A TVA OFFICIAL (1943)
BEFORE SETTING FORTH details of the working relationship between the
TVA and the land-grant colleges, a brief review of some important
considerations with regard to the agricultural extension services is
required. 1 While the experiment stations of the colleges are also in-
volved, the extension services have a central role, accounting for the
bulk of personnel and funds assigned to the TVA fertilizer and readjust-
ment programs.
CHAKACTER OF THE EXTENSION SERVICE
The network of land-grant colleges which now covers the United States
originated in the Morrill Act of 1862, which provided for a grant of
30,000 acres of land or its equivalent in scrip to the states for each
representative and senator in Congress. This established an endowment
fund, the income from which could be used for the support of agri-
cultural colleges. This income, which varied from substantial to neg-
ligible, was later increased by grants under the Second Morrill Act,
passed in 1890, and subsequent agricultural legislation. The 1890 statute
prohibited support under its provisions for colleges whose admission
requirements made distinctions on the basis of race or color, but the
establishment of separate colleges for white and colored students was
acceptable. In all, sixty-nine land-grant colleges and universities have
been established, including seventeen Negro institutions.
In 1887 Congress passed the Hatch Act, which provided for federal
support of agricultural experiment stations attached to the land-grant
colleges, on the basis of annual grants. An Office of Experiment Sta-
tions in the U. S. Department of Agriculture has exercised a rather
1 See especially Gladys Baker, The County Agent (Chicago: University of Chicago
Press, 1939) ; also Russell Lord, The Agrarian Revival: A Study of Agricultural
Extension (New York: Association for Adult Education, 1939) ; V. O. Key, Admin-
istration of Federal Grants to the States (Chicago: Public Administration Service,
1937) ; G. A. Works and Barton Morgan, The Land-Grant Colleges, Staff Study No.
10, Advisory Committee on Education (Washington: U. S. Government Printing
Office, 1939).
[117]
118 University of California Publications in Culture and Society
loose supervision, limited largely to passing upon the legality of projects,
leaving much to the initiative of the individual stations.
Agricultural extension work dates officially from the Smith-Lever
Act of 1914, providing for a cooperative Federal-State Agricultural
Extension Service for instruction and practical demonstration to per-
sons not actually resident at an agricultural college. This act laid the
basis for the county agent system, with funds contributed by counties,
states, and the federal government. Administration of the act was im-
plemented by the "1914 agreement" between the Department of Agri-
culture and most of the land-grant colleges. The Department agreed
to undertake all extension work through the agricultural colleges. Ex-
tension agents were to be joint employees of the colleges and the Depart-
ment. However, the Department did not undertake to approve the
selection of agents, and in general its Cooperative Extension office has
not exercised a close or detailed supervision. The Department's super-
visory unit is a "mere skeleton organization in comparison with the state
and county field forces." 2 The presumptively basic weapon of the De-
partment, the authority to withhold funds, has been used only once in
the case of Puerto Rico, in addition to a threat against Mississippi in
1930. In general, the state organizations follow their own programs,
receiving grants-in-aid from the federal government, with little more
than "advice and persuasion" as day-to-day administrative control from
Washington. Doubtless the existence of latent broad powers in the fed-
eral government which could be exercised if the situation warranted
the risk of political crisis induces a measure of voluntary restraint
against gross inadequacies.
The relation of the TVA to the state extension services follows the
pattern established by the Smith-Lever system. It is therefore necessary
to refer to certain characteristics of that pattern which have evolved
over the years. These characteristics include (1) indications that the
county agent is committed to "court house politics" on the county level ;
(2) the relation of the extension services to a private quasi-political
organization; (3) a tendency of the extension agents to deal primarily
with more prosperous elements of the farm population; (4) the emer-
gence of new administrative duties of county agents, associated with
changing organizational objectives of the extension service.
1. While there is some tendency for financial support of county agents
to be restricted to state and federal sources, the traditional pattern of
some contribution by the county government prevails. As a consequence,
2 Baker, op. cit., pp. 60, 102 ff.
Selznick : T VA and the Grass Roots 119
the county agent tends to be continuously involved in factional contro-
versies at the local level and especially to devote a considerable portion
of his attention and effort to insuring the continued appropriations of
county funds for the extension office. Even where the contribution of
the county is relatively small, the extension service of the college will
not make an appointment which meets opposition from the county lead-
ership. In the poorer counties of the South, the county agent looms
rather large in the local machinery, and his need for political accom-
modation is correspondingly great. The need to render special favors
to potent individuals or organizations cannot readily be ignored by an
agent who wishes to hold his job. 3 This situation is generally recognized
by extension officials in the TVA area and accounts for some of the
reluctance of assistant agents on the TVA program to be promoted to
county agent.* The special relation of the county agent to local govern-
ment emphasizes the grass-roots character of the instrumentality, but
at the same time it constitutes a force which tends to shape the TVA
program along special lines governed by the preexisting pattern of
leadership and control in the local area.
2. Local farm bureaus have been associated with the extension system
since its inception. 5 County associations of farmers were strongly sup-
ported and promoted by extension leaders and county agents as useful
vehicles for implementing the rural education program. At first, these
groups acted as the arms of the extension service, but as state and
national federations were organized, legislative matters came to have
a predominant place in the attention of the farm bureau leadership.
By the time of the organization of the American Farm Bureau Federa-
tion in 1920, the political orientation was fixed, so that what began as
an educational device became an organized pressure group. As a na-
tional machinery was established, and a corps of organizers sent out
into the farm areas, the role of the county agent remained central, but
he became increasingly dependent upon the farm bureau organization
and consequently was often committed to the policies of the national
leadership. It is true that local farm bureau organizations do not always
or necessarily agree with or carry out the policies of the Washington
office of the American Farm Bureau Federation, but their dues contri-
butions support the national organization. In a few states the county
farm bureau still has legal status as the agency cooperating with the
3 Lord, op. cit., p. 110.
4 See below, p. 140.
5 Baker, op. tit., pp. 15 ff.
120 University of California Publications in Culture and Society
extension service, which has prompted rival organizations to protest. 8
In the Southern states, the farm bureaus are not as strong as in, for
example, Iowa and Illinois, so that the local organizations exercise
rather less control over the extension service, and as Baker suggests, 7
the relation of dependence may be reversed. In Tennessee the state
director of agricultural extension is a member of the board of directors
of the Tennessee Farm Bureau Federation. This is also true in North
Carolina and other states in the region. In Alabama, in August, 1942,
it was observed that county agents had posted recruiting posters for
the farm bureau, which stressed that the American Farm Bureau Fed-
eration had procured 110 per cent of parity for the farmers. This was
at a time in which the national organization was deeply involved in
lobbying activities at the capital.
The Department of Agriculture has recognized the anomalous impli-
cations of the extension service-farm bureau relationship. In 1921, a
memorandum of understanding between the Department and the Amer-
ican Farm Bureau Federation attempted to distinguish between the
farm bureau at the county level, which might legitimately contribute
funds to extension service work, and the state and national federations
(American Farm Bureau Federation), with which there were to be
no official relationships. The agreement provided that :
The county agents will aid farming people in a broad way with reference to prob-
lems of production, marketing and formation of farm bureaus and other cooperative
organizations, but will not themselves organize farm bureaus or similar organiza-
tions, conduct membership campaigns, solicit memberships, receive dues, handle
farm bureau funds, edit and manage the farm bureau publications, manage the
business of the farm bureau, engage in commercial activities or take part in other
farm bureau activities which are outside their duties as extension agents.
The number of states having such a formal tie has been steadily declining. The
formal connection, however, is not necessarily the significant point. In some states
where the legal tie exists, it may not be meaningful. However, where the legal link
does not exist, the informal bonds may continue strong. In 1947 a spokesman of the
National Farmers Union claimed that "in more than half of the States of this Nation,
the Extension Service is not a free, educational agency, but so closely connected with
a single organization within agriculture that the two are sometimes indistinguishable,
one from the other." See Hearings on S. 1251, Senate Committee on Agriculture and
Forestry, 80th Cong., 1st sess. (1947), p. 385. Yet legally the tie is to a "farm bureau"
without reference to the American Farm Bureau Federation, so that theoretically a
local group need not affiliate with the AFBF. The Ohio Farm Bureau Federation
has often been opposed to the policies of the national organization, and even the
New York Farm Bureau Federation opposed S. 1251 (National Soil Fertility Bill)
though the bill was strongly supported by the national headquarters. Nevertheless,
the AFBF is one of the great lobbies of the United States, and its intimate relation
to the agricultural extension service is one of the striking characteristics of the
structure of American agricultural organization.
7 Baker, op. cit., p. 141.
Selznick: T VA and the Grass Roots 121
and further:
The county farm bureaus have their State and national (American) Farm Bureau
Federations, which are working on economic and legislative matters and are also
promoting the extension service and agricultural education and research. These
federations are, however, not directly connected with the extension service and
do not enter into cooperative agreements with the State colleges and the Department
of Agriculture involving the use of federation funds and the employment of exten-
sion agents, and the college and the Department are not responsible for the activities
of the farm bureau federations. 8
In fact, however, these rules have not been generally effective. Twenty
years later, it was necessary for the Secretary of Agriculture to issue
the following statement :
Eecently, reports have reached the Department of Agriculture that officers or
employees of the Department have participated actively in meetings and in other
activities concerned with the establishment of general farm organizations, or with
recruiting members for existing farm organizations.
It has long been the established policy of the Department that its officers and
employees shall refrain from taking any part in activities of this type. This is a
necessary corollary of the equally long-established policy of the Department that
it shall deal fairly with all farm organizations and deal with each upon the same basis.
As a continuation of this policy, it should be understood by all officers and em-
ployees of the Department that it is not permissible for any of them to
1. Participate in establishing any general farm organization.
2. Act as organizer for any such general farm organization, or hold any other
office therein.
3. Act as financial or business agent for any general farm organization.
4. Participate in any way in any membership campaign or other activity designed
to recruit members for any such organization.
The phrase "general farm organization" used in this memorandum is intended to
refer to such national, regional or State farm organizations as, among others, The
National Grange, The American Farm Bureau Federation, the Farmers' Union.
The Farmers' Holiday Association, The Farmers' Equity League, the Missouri
Farmers' Association, and their regional, State and local constituent groups. 9
This memorandum has apparently not been enforced with respect to
the extension service, perhaps because extension officials in the field
think of themselves as state rather than federal employees, and because
of the long-standing pattern of loose control from Washington.
3. The tendency of the county agent to deal with relatively more
prosperous farmers has received considerable comment and some study. 10
8 This agreement is reproduced as an appendix to Works and Morgan, op. cit.
9 Memorandum No. 893, Office of the Secretary, U. S. Department of Agriculture,
Washington, D.C., March 21, 1941 (mim.).
10 See Baker, op. cit., pp. 212-213. Presumably as a response to criticism suggested
by statements of this sort, a number of studies have attempted to evaluate the extent
to which the extension service reaches different strata of the population. M. C. Wilson
122 University of California Publications in Culture and Society
Attention to this problem has been increased with the intervention of
government agencies, such as the Farm Security Administration, whose
work was specifically directed toward the lower third, and the rise
of farm organizations such as the Southern Tenant Farmers Union 11
and the National Farmers' Union, which have strongly criticized
both the extension service and farm bureau system. In the Southern
states the situation is aggravated by the neglect of Negro extension
needs, but basically it appears that the nature of the extension service
tools, rather than any "Bourbon" attitude, sets the pattern. Extension
leaders often admit that their organization cannot reach all the farmers,
but they point out that their traditional function has been education.
A county agent must work with those who possess the means for under-
taking new agricultural methods, and no funds or other material aids
have been available for those who could not do so. Consequently, it was
natural that the county agent should tend to ignore those who did not
have the means to participate, and that he should build up his clientele
primarily among those who did have some resources. Home demonstra-
tion agents have faced the same problem, since poorer families could
not afford to put advice for home improvement into action. As edu-
("How and to What Extent is the Extension Service Eeaching Low-Income Farm
Families?" U. S. Department of Agriculture Extension Service Circular 375, De-
cember, 1941) concludes that on the county level "slightly more of the farm families
in the average and above-average socio-economic segments than of relatively disad-
vantaged farm families in the same areas are being reached by Extension. The dif-
ferences, however, are not great. In fact they would seem to be even smaller than
might be naturally anticipated in a situation involving voluntary participation of
people in Extension teaching activities. Extension must render reasonable service to
those progressive people who already have the desire for information. This fre-
quently limits the attention a limited county staff can devote to arousing a desire
for information on the part of those who do not yet have that desire." The term
"reached" used here is very important, and it is used in Wilson's report to mean
"in some measurable way." However, D. L. Gibson ("The Clientele of the Agricul-
tural Extension Service," Michigan Agricultural Experiment Station Quarterly Bul-
letin, Vol. 26, May, 1944) analyzes the problem without leaving the term "reached"
ambiguous. Gibson finds that in the Michigan counties studied the differences among
socioeconomic groups were not great in respect to such passive participation as re-
ceiving mimeographed material, but the differences were considerable when more
active types of participation, such as personal visits and running test plots, were
considered. Gibson concludes that, for the area studied, "the so-called 'disadvantaged
class' of farmers, those of lower socioeconomic status, participated decidedly less in
the agricultural extension program than did those of high status." As Gibson sug-
gests, this is no reflection upon the efforts of county agricultural agents. It is im-
portant objectively, however, in considering the character of the extension service.
See also C. E. Hoffer, "Selected Social Factors Affecting Participation of Farmers in
Agricultural Extension Work," Michigan State College Agricultural Experiment Sta-
tion Special Bulletin 331 (June, 1944) ; and W. A. Anderson, "Farm Youth in the
4-H Club," Department of Rural Sociology, Mimeographed Bulletin No. 14 (Ithaca,
New York: Cornell University Agricultural Experiment Station, May, 1944).
"Recently renamed, National Farm Labor Union, affiliated with the American
Federation of Labor.
Selznick: T VA and the Grass Roots 123
cators, the agents have been under pressure to make a good showing
so that they often turned to the more well-to-do farm families who might
be used to advantage in reporting accomplishments.
Extension service officials in the South tend to accept the prevailing
code with respect to farm tenancy. Thus one farm management special-
ist in Georgia pointed out that the basis for criticism of the extension
service was the tenancy situation. But in extenuation he said, "The
extension service must deal with the landlords, but in this way we do
reach the 'croppers'." The proportion of farm tenancy is very high,
so that the extension service may, in this state, deal directly with only
the upper third of the farmers. With tenants having little or no material
means of their own, and being highly mobile, it is not surprising that
the county agents should restrict their attention to the more substantial
farming population, which can participate in a long-time educational
program.
4. The role of the county agent as an itinerant teacher, an outpost
of the agricultural college, has been subject to considerable changes.
Even during the first world war, the county agent played an adminis-
trative role in attempting to push a national program of agricultural
production. The slow method of demonstration gave way to a series of
campaigns, organized around concrete objectives such as crop quotas.
In addition, problems of farm labor and food conservation were added
to the wartime functions of the local agents, to say nothing of war bond
drives and miscellaneous services. However, the basic orientation of
extension as an educational rather than an action organization persisted.
Yet the administrative potential of a field force thus widely spread over
the country could not be overlooked, and when depressed economic con-
ditions brought large-scale government intervention for the benefit of
farmers, the educational aspect of the system was subjected to increased
attenuation. Some of the stronger colleges resisted the new tendency,
but those with little available funds, especially in the South, welcomed
the new programs. Perhaps the most important of these was the work
of the Agricultural Adjustment Administration, which provided funds
for the employment of numerous additional county agents and assist-
ants. The county agent assumed a central administrative role and was,
as a rule, deeply involved in many different action programs. Hence
the character of the extension service underwent a fundamental change
from that which the Smith-Lever Act envisioned. Far from being merely
a teacher, devoted, at least in theory, essentially to noncontroversial
promotion of better farming methods, the county agent became an exec-
124 University of California Publications in Culture and Society
utive with commitments and involvements which forced him into a
strategic position in the farm leadership of his community.
These characteristics of the extension service have been reviewed
only in order to make clear some of the factors which might be expected
to shape the relationship of the Tennessee Valley Authority to the state
institutions. In the interaction between these organizations, the charac-
ter of each represents the effective channel as distinct from the formal
relationships through which the proposals of policy become disposed
in action.
PATTERN OF COOPERATION
The cooperative arrangements between TVA and an individual state
university have usually been embodied in a master contract for a "joint
program of agricultural development and watershed protection through
improved fertilization," signed by the president of the college or uni-
versity and the Chairman of TYA. Statements of purpose and policy
are laid down, centering around the encouragement of rational farm
management practices which would afford maximum benefit to the
farmers from the use of new fertilizer products, all within a "regional
program of watershed protection and agricultural development by which
soil erosion and flood waters may be controlled and soil fertility may
be restored and permanently maintained." 12 Provision is made for a
broad cooperative program involving soil surveys in Tennessee Valley
counties, laboratory and field research to be carried on at agricultural
experiment stations, and farm unit and area demonstrations using the
facilities of the agricultural extension service. In general, the actual
work is carried on by the college organization, the Authority reimburs-
ing the institution for its expenditures on additional personnel and
equipment. 13 Facilities and equipment acquired under this arrange-
ment became the property of the Authority. The master contract is
implemented by a series of project agreements which are signed by the
TVA's Director of Agricultural Relations and the head of the experi-
ment station or extension service within the college.
These formal agreements are made within a context of loose and tra-
ditional ties and, as written instruments, have not always been strictly
observed or even kept up to date. While some criticism on this score
has arisen within the Authority, in fact the ability to conduct the pro-
12 From a sample contract.
13 Reimbursements are made for detailed items. The possibility of making lump-
sum grants to the colleges for executing a delegated responsibility has been explored,
but rejected. The states need not match funds provided by the Authority, though
such a procedure was included in the original Smith-Lever Act of 1914. See n. 16,
below.
Selznick : T VA and the Grass Roots 125
gram on an informal basis is deemed an asset. The grant-in-aid proce-
dure is dominant, with the traditional Smith-Lever pattern 14 consciously
maintained. In some respects, the states have an even greater advantage
in dealing with TVA than they do with the Office of Cooperative Ex-
tension in the USDA. Thus the dean of one of the agricultural col-
leges in the Valley pointed out that the colleges are not limited to a
rigid budget in respect to grants from the TVA, but can return to
the Authority for interim revisions of project agreements allowing for
increased funds during a given year. The usual role of the TVA agri-
culturists has been characterized as "passive and advisory," borne out
by their persistent emphasis upon the need for programmatic direction
to come from the states.
The organizational implications of this pattern of cooperation come
in part from the collateral objectives which define the methods avail-
able to the Authority for administering its program. These qualifying
conditions, self-imposed, concretize the grass-roots policy, presumably
offering criteria for measuring the extent to which it has been applied
or been successful. Cooperative projects should (1) stimulate the local
institutions to an interest of their own in the joint activities, so that
they may gradually increase their own financial contributions and
ultimately even carry on independently if support by TVA were with-
drawn; (2) have a leavening effect on the regular extension service pro-
gram; and (3) broaden the perspective of personnel in the Authority.
These criteria and objectives sustain and guarantee the traditionally
loose relationship between the land-grant college organizations and
the federal government. 13 Such an interpretation is warranted by the
following considerations :
1. The desire to stimulate the local institutions tends to foster rela-
tionships which are informal and flexible, thus minimizing business-
like quid pro quo attitudes which an emphasis on reciprocal contractual
obligations may induce. The colleges may be nursed along and permitted
to deviate from formalized procedures, especially if there is the possi-
bility that added responsibilities will be accepted. An unusual amount
of consideration and attention of the TVA agricultural staff is devoted
to preserving the integrity of the colleges so that they may receive full
11 The pattern developed between the Department of Agriculture in Washington
and the land-grant colleges in administering the Smith-Lever Act of 1914 under
which agricultural extension work was inaugurated on a national scale.
13 It will probably bear reemphasizing that this study is not concerned with apprais-
ing the grass-roots method in terms of its formal objectives ; interest is restricted to
the organizational implications which throw light on the interaction of the Authority
and the colleges.
126 University of California Publications in Culture and Society
public credit for participation in the cooperative program, and particu-
larly to supporting the colleges as indispensable channels to the farm
population. At the same time, little systematic attempt is made by the
Authority to gauge the extent of contributions by the institutions. While
the grass-roots approach has often been discussed and justified, no clear
evidence of a significant increase in contributions, especially financial,
by the colleges to the joint program has been presented. Even as late
as 1943, attempts to make estimates along this line were faced with
considerable difficulty. There is, moreover, some doubt as to whether
all funds contributed by the colleges are derived from local sources.
Grants-in-aid from the U. S. Department of Agriculture are thought
of by the colleges as their own funds, and the TVA does not question
the sources of funds applied by the institutions to a particular project.
Thus there is the possibility that resources used by the institutions may
represent actually a federal rather than a state contribution. Lack of
clarity on this problem appears to be due to the conviction of the TVA
agriculturists that the grass-roots method has an inherent validity which
does not require any objective tests. Indeed, the colleges are reluctant
to provide information as to their financial contribution to cooperative
projects, and one TVA staff member reported that the Director of Agri-
cultural Eelations has instructed his department not to request such
information from the colleges, "since they don't want to give it." 16
2. The objective of influencing the regular program of the experiment
stations and extension services involves the corollary procedure of inte-
grating the TVA program with that previously developed within the
college organizations. The articulation of the TVA and extension serv-
ices program will be described below. 17 At this point it suffices to suggest
that the TVA comes to have a special interest in and commitment to
the regular program of the extension services so that, to some extent,
the institution can carry out its obligations under the joint program
simply by continuing previous services to the farm community. Per-
sonnel assigned to TVA work may do other extension work without fear
18 This does not mean, of course, that stimulation has not occurred. On the contrary,
there is evidence that in some fields the colleges have expanded on their own from the
core represented by the TVA-subsidized program. It is not known to what extent
some other method, including lump-sum grants and matching provisions, would have
affected the colleges' initiative. The point here is simply that the objective meaning
of the doctrine of stimulation sustains a loose administrative relationship. It is note-
worthy that reimbursement for detailed items, while superficially representing a
measure of control by the Authority, is in fact more congenial to those who advocate
a loose relationship than is the lump-sum grant. The use of the latter procedure im-
mediately raises the question of a system of fiscal control and work planning, as well
as the possibility of instituting a matching requirement.
17 P. 129 ff.
Selznick : T VA and the Grass Roots 127
of being questioned by TVA, thus underlining in action the continuity
of TVA test-demonstration projects with customary extension proce-
dures. Furthermore, the integration of the two programs vitiates efforts
to measure contributions from the institutions, so that relationships are
based fundamentally on a faith in the logical indivisibility of extension
and TVA activities. The extension service has in fact been influenced
by the TVA program, especially in being provided with a means (ferti-
lizer) of material aid to induce farmer participation, and in having a
well-defined set of objectives on which effort might be concentrated.
However, the very process by which the extension service absorbs TVA
fertilizer distribution and testing activities demands that the extension
organizations be permitted to be free of over-diligent supervision or
inspection from the TVA offices. This has been recognized by the TVA
agriculturists who have consciously refrained, for example, from send-
ing inspection teams into the field to check on the work of the extension
service.
3. It is held desirable that TVA personnel will find new and broader
perspectives if they consciously attempt to exploit the resources avail-
able in the local institutions. This doctrine objectively strengthens the
relative weight of the local institutions in their relations with TVA
insofar as adaptation to the local organizations (and to the grass-roots
theory generally) on the part of TVA personnel elicits official approba-
tion. Far from approaching the local agencies with a "reformer com-
plex," this viewpoint assigns a special value to those agencies and exalts
them as a source of education and inspiration. However, in the agri-
cultural program, which is administered by TVA officials who come
initially from the colleges, it appears doubtful that any significant
education of this sort occurs. Since the TVA agriculturists are them-
selves products of the extension service system, and among the most
vigorous advocates of that system, it is not easy to see how they them-
selves are broadened by contact with the colleges. However, it is no
doubt true that other divisions of the Authority, having different
origins, are shaped to some extent by working with the local organiza-
tions. In any case, this doctrine of exploiting local resources tends to
justify and bolster any influence which the college staffs may have upon
TVA personnel.
The strong belief of the TVA agriculturists in the land-grant college
system is, as a set of organized sentiments, the fundamental basis of
the administrative relationships. This, as has been observed, results from
the institutional origins of the leading personnel and from the traditional
128 University of California Publications in Culture and Society
precedents laid down in the administration of the Smith-Lever Act, as
well as from a defensive psychology of representatives of the region.
These sentiments are sustained by the grass-roots doctrine, which in-
volves an essentially localist conception of democracy. In addition, an
administrative logic of working through the land-grant colleges has
been developed, asserting that democratic administrative policy leads
inexorably to the pattern of cooperation already laid down. This logic
maintains that the TVA fertilizer-testing and soil conservation pro-
grams would not have been welcomed by the individual farmer as
isolated practices to be appended to his existing plan of work; they
require rather a total reorganization of the individual farm plan. "But,"
runs the argument, "for these conservation and fertilizer practices to
be integrated with other farm problems and presented to the farmer
as a unified program they would require either that the Authority
establish an agricultural service which would seek to carry out a com-
plete program of research ranging from purest research to the giving
of advice to the farmer on his farm and the farmer's wife in her kitchen,
or that the Authority fit its special interests into existing farm pro-
grams." If the Authority had elected to create its own agricultural
service, it would have had to undertake an educational program at the
county and community level which would be in line with the character
of the county agent system. Moreover, its field agents would require
training, and this would involve the establishment of agricultural col-
leges, including experiment stations for the controlled testing of farm
practices. In the end, therefore, if the TVA had decided to set out on
its own, its organization would have been "patterned after the ideals
of the land-grant colleges." It therefore makes eminent good sense,
in this view, that the TVA should have chosen to work through the
existing college system.
It seems clear that this all too innocent formula reflects the funda-
mental conviction of the TVA agriculturists that there is an identity
which unites the TVA program in agriculture and that of the land-
grant colleges. Cooperative arrangements with the colleges would
inevitably have been undertaken by any federal agency engaged
in soil conservation, but the specific pattern of cooperation developed
by TVA cannot be adequately understood apart from the ideological
bent and institutional commitments of the TVA agriculturists.
The implications of the administrative logic just paraphrased go be-
yond the relations of the colleges to the TVA. By a similar logic, it
would seem that the newer agricultural field agencies, such as Soil
Selznick : T VA and the Grass Roots 129
Conservation Service and Farm Security Administration, cannot or
ought not remain side by side with the colleges, since they tend to
converge in function. To the extent, therefore, that there is a struggle
among these agencies, the TVA arguments throw the weight of the
Authority on the side of the land-grant institutions. To avoid dupli-
cation, either the new agencies would have to be absorbed by the
colleges, as in the case of the TVA agricultural program, or a drastic
reorganization of the extension service would have to be undertaken,
divorcing the administration of agricultural action programs from the
educational function. The latter alternative would put a wedge between
the colleges and their client publics, hobbling their influence. But it
is precisely this access to the farmers which is at stake, a prerogative
which was amply protected in the pattern of cooperation won by the
colleges in their relation to the TVA. Of course, the administrative
logic defended in the TVA disregards other agencies that have chosen
to build field forces to reach the farmer with a special program. The
sheer convergence of formal function (as of education, or of the pres-
entation of an integrated farm program) does not seem to weigh heavily
in the minds of those who feel that the traditions and commitments of
existing institutions will not permit adequate administration of pro-
grams which require a special loyalty, vision, or zeal.
INTEGRATION OF TVA AND EXTENSION-SERVICE
PROGRAMS
The joint agricultural program of TVA and the land-grant colleges of
the Tennessee Valley includes (1) the controlled testing of plant food
products manufactured by the Authority according to standing pro-
cedures of the agricultural experiment stations, including work in
laboratories, greenhouses, and on field plots; and (2) a program of
large-scale testing and demonstration of phosphatic fertilizers on actual
farms. The latter is known as the "test-demonstration" program and,
carried on through the extension services, represents the major phase
of the work. 18 In addition, a program of assistance to farm people and
rural communities affected by the flooding of farm lands by TVA reser-
voirs is also carried on through the local extension-service organizations.
Since the activities of the extension service in the joint program rep-
resent the TVA channel to the farm people of the Valley area, and
constitute overwhelmingly the largest part of the TVA effort in agri-
culture, they assume a primary importance in the grass-roots approach.
18 See above, pp. 99 ff., for data on extent of the program.
130 University of California Publications in Culture and Society
Cooperative agricultural extension work was defined in the Smith-
Lever Act of 1914 as "the giving of instruction and practical demon-
strations in agriculture and home economics to persons not attending
or resident in said colleges in the several communities, and imparting
to such persons information on said subjects through field demonstra-
tions, publications and otherwise," The joint TVA-extension-service
program is basically a continuation and intensification of agricultural
extension work, with, however, some modifications necessitated by the
special objectives of distributing TVA fertilizer and carrying on the
relocation and readjustment work. The TVA Act authorized the Au-
thority to arrange with farmers and farm organizations for use of new
forms of fertilizers under conditions permitting an accurate measure
of the economic return they produce. This qualification added a new
factor to the extension-service work, for while demonstration had long
been a cornerstone of agricultural extension, testing had been left to the
controlled, experimental investigations of the stations. Consequently,
the TVA adopted the idea of a test-demonstration program, wherein
the scientific function of appraising the practical value of concentrated
fertilizers would be combined with the educational function of encour-
aging individual farmers and their neighbors to try the new materials
and spread their use.
In 1935, after a year of work devoted to testing Authority fertilizers
at the experiment stations, the test-demonstration program was in-
augurated. Broadly conceived as a contribution to the total agricul-
tural problem of the region, this program of distributing phosphatic
fertilizers was to serve a dual objective : water control on the land in
conjunction with the adoption of methods of farming beneficial to the
welfare and security of farm people in the area. Phosphates are integral
to a system of farming which emphasizes the growing of leguminous
forage plants or cover crops. In this way the soil is enriched not only
by the application of phosphate and lime but by the fixation of nitrogen
from the air. At the same time, the legumes (such as alfalfa, lespedeza,
red clover) hold down the soil and prevent the erosion which follows
upon the planting of such clean-tilled crops as cotton and corn, espe-
cially in hilly regions. Moreover, a shift from a corn-and-cotton economy
to one which places a greater emphasis on livestock is encouraged ; this
has long been an objective of agricultural agencies in the South. Hence
the obligations of the TVA to protect its reservoirs and decrease the
danger of floods, as well as its responsibilities for fertilizer manufacture
and distribution, are brought together in a program readily articulated
Selznick : T VA and the Grass Roots 131
with the educational objectives and machinery of the agricultural
extension services.
The fertilizer distribution program utilizes two basic devices, indi-
vidual farm test-demonstrations and area test-demonstrations. The work
with farms brought into the program through these devices represents
the basic and dominant phase of the joint activities. The machinery of
the extension services is used exclusively, 19 the TVA reimbursing the
colleges for the employment of assistant county agents who, working out
of the county extension offices, are assigned to the test-demonstration
work. In addition to the assistant county agents, supervisors and
other specialists are employed by the colleges with salaries reimbursed
by TVA. These officials of the extension service constitute the operating
organization which brings the TVA fertilizer and soil conservation pro-
gram to the rural population.
In theory, and to a varying extent in practice, farmers on the program
are selected by the communities themselves, through meetings called
by the county agent. The distribution of fertilizer is handled by a
county soil erosion control association to which bulk consignments are
made by the Authority. However, there is no question that the county
agent (or the assistant county agent assigned to the program) is the
key man. It is "his" program and "his" association, and it is under-
stood in practice that all contact with the farmers and the farm organi-
zations which implement the program will be made through him. 20 The
selection of farmers to participate in the program is in practice the
responsibility of the extension service official, though this is modified
to some extent by the use of community committees and meetings.
Sometimes the already established AAA committees were used to in-
augurate the test-demonstration program.
The unit test-demonstrators usually enter the program for a period
of five years. Each agrees to commit his entire farm to the testing pro-
gram, involving the reorganization of the farm plan to increase the
production of food for the family and feed for the livestock; to re-
adjust land-use practices so as to halt erosion and build up the soil;
to use the phosphatic fertilizers provided by the TVA, combined with
lime, to increase the acreage of cover crops; to leave check plots un-
treated for comparative purposes; to pay .freight charges on shipment
of fertilizers ; to keep records and provide access to the farm by exten-
sion service personnel and other farmers. These contributions by the
19 See table 4, p. 102.
30 A more detailed discussion of the county soil associations will be found below,
pp. 230 ff .
132 University of California Publications in Culture and Society
farmer are strongly emphasized, in order to dispel the idea that free
fertilizer is being distributed.
The area test-demonstrations involve the distribution of fertilizer to
an entire watershed community. At the initiative of the county agent,
a community organization is formed, and a survey of the area and in-
ventory of its resources is made as a basis for the plan under which
each farmer will participate in the soil conservation program, receiving
TVA phosphate as a material aid. An attempt is made to hold regular
monthly meetings at which farming plans and other community prob-
lems are discussed.
In Tennessee, in 1942, there was one farm unit test-demonstration for
each fifty farms in the Tennessee Valley counties of the State; area
test-demonstrations included about four per cent of all farms. It was
calculated that costs to the public per test-demonstration farm (unit
and area) amounted to $100 a year, one-half allocated to the cost of
the phosphates and the other half to salaries and travel of extension
service personnel. The farmers themselves are believed to have invested
about an equal amount in lime, terraces, and other improvements.
A basic procedural principle which runs through the entire admin-
istration of the test-demonstration program is the goal of integrating
that program within, and thereby strengthening, the county agricul-
tural extension program. Each step in that direction, each evidence that
the extension service has absorbed the TVA program into its own, is
considered an achievement. In the relocation and readjustment program
as well, although dealing with problems of an extraordinary character,
the ultimate objective is the attainment of a stage which will permit
the job to become part of the regular program of county extension work.
The initial phase of aid to reservoir-affected farm families and rural
communities is that of relocation. A series of conferences involving the
Authority and Extension service officials, extension workers, county
program planning committees, and subordinate county relocation com-
mittees, down to neighborhood groups meeting with county agents, are
held. Detailed surveys of the area are made. Lists of farms for sale or
rent are compiled and appraisals made by extension officials. Informa-
tion and more material aid (such as in moving) are made available
until the flooded-out families have removed to their new homes or left
the area. Steps in the process of population removal and readjustment
have been formulated as :
1. Study and analysis of the social and economic conditions of the region affected by
reservoir flooding.
Selznick : T VA and the Grass Roots 133
2. An appraisal of all the resources available for readjustment.
3. Formulation of plans for the evacuation of the reservoir area and the readjustment
of the displaced families.
4. Visitation of all families living within the reservoir area.
5. The removal of families from the reservoir area.
6. Eeadjustment of families at their new locations.
During the relocation phase, the Keservoir Property Management
Department of the Authority takes an active part, with the extension
service the main cooperating agency. Up to April, 1943, thirteen reser-
voir areas had been evacuated, displacing 9,489 families, in addition
to 1,819 families removed from areas of projects then under construc-
tion. 21 The Keservoir Property Management Department reported that
only five families were evicted by court order, all from the Norris area
(the first to be evacuated) where plans had not been put into action
before the rising of the lake. Farm ownership has varied from 59 per
cent in the Norris area to 7 per cent at Wheeler dam, averaging 35 per
cent of the total families displaced. Less than 6 per cent of the families
removed up to 1943 were Negroes.
As the process of relocation is completed, the readjustment phase
begins. It is here that close integration with the regular county exten-
sion program is made possible. Additional assistant county agents and
supervisors added to the extension staffs for the readjustment program
(with salaries reimbursed by TVA) devote themselves to aiding farmers
to achieve a successful adjustment in their new locations. The test-
demonstration program, especially the area watershed device, has been
used as a vehicle for bringing educational and material aid to these
farmers. During this phase, the duration of which is not defined, the
test-demonstration program is extended on the basis of the needs of
the relocated farmers. Consequently the same articulation with the
extension service obtains in the readjustment program as in the test-
demonstration work generally. The assistant agents assigned to readjust-
ment work may, in addition, be called on to assist in the administration
of the rental of Authority-owned lands through the land-use associa-
tions set up in the reservoir areas.
With this brief sketch of the program as background, we may now
turn to a consideration of some of the problems raised by the TVA
channeling its agricultural activities through the extension service
machinery. It will be evident that the discussion above of the character
of the extension service 22 underlies the following analysis.
- 1 See table 7 above, p. 104.
22 Pp. 117-124 above.
134 University of California Publications in Culture and Society
1. The "testing" dilemma. Section 5 (b) of the TVA Act authorizes
the Authority "to arrange with farmers and farm organizations for
large-scale practical use of the new forms of fertilizers under conditions
permitting an accurate measure of the economic return they produce."
This authorization is the basis of the test-demonstration program, which
in turn represents the core of TVA agricultural activities. Yet, in the
administration of the program, there appears to be a constant straining
of the actual operations against the official interpretation. The county
agent system is not a research organization and its personnel in the
counties are not much concerned over problems of scientific appraisal.
It is sometimes said that extension personnel tend to be evangelistic
rather than scientific in outlook. They conceive of their job as educa-
tional, as it traditionally has been, and they are more interested in the
demonstration than in the testing aspect of the fertilizer program. Once
controlled tests are made at the experiment stations, these agents con-
ceive of the value of the concentrated phosphatic fertilizers as estab-
lished, the job being to prove to the farmers in the local areas that they
are indeed useful and profitable. In general, the extension agents use
the TVA fertilizer as a tool in developing the general agricultural ex-
tension program in the county. This, being in line with the conscious
policy of integrating the program with that of the extension service, is
approved by the TVA. But at the same time the possibility of isolating
the effect of using TVA phosphates is reduced, thus undermining the
testing function.
The extension service program is broadly educational, related to the
general development and welfare of the agricultural community and
its resources. The statutory authority of the TVA, however, is confined
to the limited problem of fertilizer manufacture, testing, and distribu-
tion. But not only is the TVA organization actually interested in pro-
moting the same program as that of the extension service, but it is
often said by the Authority's agriculturists that "the TVA has no
program of its own in agriculture." The distribution of TVA fertilizers
becomes in fact a material aid to be used in advancing the extension
service program. As a result, the TVA personnel tend to be unclear
and sometimes inconsistent in interpreting their program. The possi-
bilities of attaining an accurate measure of the results of using phos-
phates on a farm will be dissipated as more factors are added which
affect the economic return of the farm unit. But not only does the
extension service, through mustering educational resources in helping
the farmer, add to the factors, but there are also attempts with TVA
SelznicJc : T VA and the Grass Roots 135
support to include the test-demonstration farms in a rural electrifica-
tion program and to concentrate farm forestry work on the same farms.
At the same time, it is insisted that the farmers are testing the value
of TVA phosphates.
This difficulty is recognized by the Agricultural Relations personnel,
and there has been some suggestion that the testing and demonstration
functions should be separated, frankly recognizing that the county agent
staff is concerned exclusively with the latter. Moreover, there is some
tendency to move to an emphasis on the area demonstration device,
which is much less closely specified or controlled than the individual
farm demonstration, and is justified partly on a somewhat different basis.
The mass distribution of fertilizers to an entire community is viewed
as a contribution to TVA's flood-control and dam-protection responsi-
bilities (holding down the soil and storing water in the soil through
cover crops) rather than as a use of the narrow fertilizer testing author-
ity. That the idea of testing is partly ideological is shown by the emphases
upon it when convenient, and its discard when inconvenient in terms
of the real objectives of the program. Thus fertilizer has been refused to
a group-farm project of the Farm Security Administration on the
ground that it was an unrepresentative type of farming, but fertilizer
has been used on equally unrepresentative types of farms when these
have been extension service projects.
2. The representative farm. The contracts with the colleges specify
that the test-demonstration program shall be based on farms of repre-
sentative size, soil variety, and type of farming. However, here too some
inconsistency seems to be induced by the relation to the extension service,
though there has been considerable effort to make it a real criterion in
selection of farms. When, early in the program, one member of the
Authority staff not an agriculturist raised the question of using
the test-demonstration program to reach progressively lower economic
layers of farm families, the Director of Agricultural Relations coun-
tered by pointing out that it was necessary to get "a representative
sample" of farms in the area in order to fulfill the testing function.
During the same period, however, county agents in the field were recom-
mending farmers who had worked closely with them for the "free phos-
phate," usually the more substantial farmers in the community. This
tendency would appear to have been a predictable consequence of chan-
neling the program through the extension service. The Agricultural
Relations Department has had to use pressure in order to get enough
small farms on the program to avoid criticism, and even this has not
136 University of California Publications in Culture and Society
been accomplished. The problem posed by the attitudes and traditions
of extension service personnel is highlighted by the following statement,
made in 1935, by a supervisor on the test-demonstration program.
In spite of the possibility of being thought obstinate, I am going to repeat that I
think you folks there are entirely wrong in your idea that small farms should be
included in the demonstrations in the hilly grazing counties of Virginia. We have
been telling these farmers for years that livestock farming was the only type of farm-
ing that could be expected to permanently succeed in this section and have further
deluged them with bulletins, farm management records, and speeches in which they
were told that farms of less than 160 acres in livestock sections could rarely be oper-
ated efficiently. ... I think if we try to use small farms for demonstration in this
area . . . our books will always show a loss unless we defeat the purposes of soil con-
servation by encouraging the growing of crops on these farms that would necessitate
frequent plowing and cultivation. . . .
The farmers who live on the small farms in this area are mostly ignorant and inef-
ficient and are not capable of conducting a demonstration for the benefit of their
neighbors. Just yesterday I was told that . . . the small farmers were suspicious of
the demonstration idea and that it would probably take considerable educational
work to get even a few of them interested. . . .
With the TVA wedded to the extension service machinery, it is not sur-
prising that the program should have been weighted by farmers selected
in terms of local commitments of the county agents. The statement of
another field agent describes the typical extension service problem in
which the available tools determine the extent of participation ; the state-
ment is in explanation of why the average size of farms on the test-
demonstration program in 1936 in his area was overly large :
Harlan is a mountain county. It is in the heart of the greatest coal mining industry
in our state. The hills are high, and the valleys generally not more than 100 feet across,
making the choice of land for fertilizer demonstration purposes quite different from
that in the broad valley of the TVA drainage basin. This, to my mind, is an instance
where theory and practice lock horns, resulting in the practical side winning, be-
cause of the fact that the men with the larger holdings have become demonstrators
because it is more feasible for them to sow a cover crop of small grain, and show
an economic return. These men are most apt to have teams and tools to prepare the
ground; they are most apt to have the cash to pay the freight on the fertilizer; they
are most apt to be able to buy grasa and legume seed to sow in the small grain next
spring, in that general territory. The man who lives at the mouth of the creek, where
generally the bottom land is widest, is usually the man who dominates the thinking
of nearly all the people who live up the hollow behind him. His word alone will usually
determine whether or not the other folk take up a new practice. . . .
Even a sincere desire to work with small farms cannot prevail against
factors which determine, for extension service personnel, the possibili-
ties of doing a job which will produce favorable results. As a conse-
Selznick : T VA and the Grass Roots 137
quence, the TVA program tends to be shaped by the same factors. A
program which emphasizes the increase of pasture lands and livestock
finds it difficult to avoid concentration on those farms which can make a
measurable contribution.
In 1947, TVA reported test-demonstration farms distributed with
respect to size : a
Under 50 acres 12.7 per cent
50-99 acres 26.4 per cent
100-179 acres 30.5 per cent
180-379 acres 23.5 per cent
Over 380 acres 7.1 per cent
Although comparisons with census figures are not conclusive because of
the prevalence of part-time farming among those holding small acreages,
this distribution seems to indicate a somewhat disproportionate partici-
pation of moderately large farms, given the nature of farming in the
Tennessee Valley area. Some indication of relative proportions which
would be acceptable in a representative sample is derived from the
specifications originally devised for selecting a sample of unit test-
demonstration farms for intensive analysis in connection with a TVA
and Tennessee-extension service progress report. Out of 61 sample farms
sought, it was thought desirable to have 27 under 50 acres, 17 from
50-100 acres, and 17 over 100 acres. The actual sample finally selected
included only 6 farms below 50 acres in size, because not enough small
farms on the program met the other specifications for the sample, includ-
ing length of time as participants. 24
In general, the unit test-demonstration farms, representing the core
of the program until at least 1940, have been disproportionately large.
Thus, in the Tennessee 50-f arm study just mentioned, the sample farms
"averaged nearly 60 per cent larger in total acreages than the average
East Tennessee farm (including part-time farmers) and had nearly
twice as many acres of cleared land." This sample, moreover, was de-
scribed as more nearly representative of farming in the area than the
entire group of unit test-demonstration farms which had been on the
program more than six years. In 1940, the farm-unit test-demonstrations
in the Tennessee Valley area of Tennessee averaged 213 acres, while
23 Hearings on S. 1251, Senate Committee on Agriculture and Forestry, 80th Cong.,
Istsess. (1947), p. 297.
34 "Progress and Possibilities for Further Progress on 50 Unit Test-Demonstration
Farms in the Valley of East Tennessee, A Preliminary Beport," mimeographed re-
port issued by the University of Tennessee (June, 1942).
138 University of California Publications in Culture and Society
according to the University of Tennessee the average size of farms in
that area was 79 acres in 1939. However, the program of area test-
demonstrations, involving the distribution of fertilizer to farmers in an
entire watershed community, has operated to decrease the average size
of farms receiving TVA fertilizers. It appears that where county agents
have exercised discretion in selection, as in the farm-unit participants,
they have leaned toward the larger farms. When, however, the com-
munity is taken as the unit of action, the program reaches broader layers.
There is no implication here that the TVA agriculturists wanted un-
representative participation ; on the contrary, they were concerned about
the problem and applied pressure upon extension personnel to rectify the
situation. As in extension work generally, the means available, and
the objectives of the program tend to select out those farmers most
capable of contributing to and benefiting from it. Moreover, this pat-
tern becomes institutionalized, including the creation of a supporting
set of attitudes, within the extension service. Consequently, the chan-
neling of the TVA program through the extension service created an
initial presumption that this pattern would be continued in the test-
demonstration work.
3. Farm tenancy. The educational activities of the extension services
tend to be largely oriented toward farm owners. The man who owns his
farm can plan his farm management and has a permanent perspective.
These conditions make him a desirable client for an educational organi-
zation. In the TVA test-demonstration program, the same factors hold,
and are emphasized by the established extension service pattern through
which it is channeled. A farm-unit participant agrees to enter the pro-
gram for a five-year period, and this alone must eliminate many tenants.
The approach in extension service and TVA has been to deal with ten-
ancy through the farm owners, considering sharecroppers as part of the
larger farm unit. While many tenants are doubtless receiving TVA fer-
tilizer through the area demonstration work, no special attention is paid
to the tenancy problem within the framework of the test-demonstration
program. 25 In fact, there is evidence that the Agricultural Relations
Department does not consider high or increasing rates of tenancy a
problem.
In the relocation and readjustment program, the problem of tenancy
cannot be ignored since the majority of farmers removed from the reser-
voir areas have been tenants (see table 7). Yet the TVA has taken the
25 The decline in effectiveness of the test-demonstration program in high tenancy
areas was attested to by a district agent of the Tennessee Extension Service.
Selznick : T VA and the Grass Roots 139
view that only the extension services have facilities for dealing with the
readjustment problem. The logic runs as follows :
The TVA has no facility for advising land owners who have sold their property to
the TVA in the matter of purchasing other farm lands. The Authority is not in a
position to know what lands are available and whether or not the price for which the
lands are offered is reasonable. The Authority is furthermore not in a position to
advise land owners who have sold property to the Authority whether or not they
should reinvest in land in their county of present residence or whether they should
invest in lands further afield. The Authority furthermore has no facility available
for agricultural advice, agricultural assistance, or agricultural education for both
land owners and tenants displaced by its land purchase program. It furthermore does
not organize 4-H clubs and does not offer any other benefits that have been performed
by the Extension Service.
While TVA has given some material assistance in the actual process of
relocation of tenant families, the readjustment phase has been under-
taken by the extension services, with funds supplied by the Authority.
In general, the approach to the tenant problem is summed up in the
following extract from a project plan for relocation of families displaced
at two of the Authority dams :
A special effort will be made to find locations for landowners who are willing to pro-
vide locations for tenants and croppers now residing on their property or on neighbor-
ing property and wishing to continue the personal tie now existing between them and
the tenant or cropper. The Authority will give preference of employment in areas
surrounding the dams to tenants, croppers and small landowners recommended by
the Extension Service as most deserving of an opportunity for accumulating capital
to be used in meeting their relocation problems. Such an arrangement will be kept
confidential.
There has been some indication that landlords in areas to which reser-
voir-displaced families have moved have been able to weed out their
own poorer tenants and replace them with better ones from the reservoir
area, adding to the problem of the most depressed layers of the rural
population. In situations of this sort, the relocation of tenants from the
reservoir sites might be accomplished, and yet indirect consequences for
the rural community might remain uncontrolled.
Neither the TVA nor the extension service appears to have handled
the readjustment program in relation to general population problems
in the region. In 1934, it was suggested within the Authority that a cor-
poration be organized to purchase from holders of foreclosed farms
options on large tracts of fertile land which could then be sold to dis-
placed farmers on a credit basis. Nothing was done, however. In 1940, a
sociological study group reported, following a brief inspection of the
program, their "dismay at the insulation of the relocation program from
140 University of California Publications in Culture and Society
that of the Resettlement Administration insulation that prevented
even referral of sharecroppers and tenants who were forced from reser-
voir property, while extensive services and follow-up were given to farm
owners who were forced to leave." This insulation, while not complete,
arose in part from the organizational hostility of the extension services
to what they considered rival farm agencies. The same group made the
following statement :
In our observations, we discovered a tendency to depreciate the few belated sociologi-
cal studies that were conducted. Some of the studies are not available for groups
such as ours or those at state universities (e.g., studies of relocated families), and
the TVA staff itself is unfamiliar with most of these sociological studies how dif-
ferent from their knowledge of researches in other fields ! We could not discover an
instance in which any of the results were utilized (e.g., a study of the Governmental
Eesearch Division revealed that 95 per cent of the people in Wheeler Reservoir area
were tenants, yet, so far as we could discover, relocation assistance and follow-up
concerned itself only with those farms which were purchased for reservoir property).
On the whole, it appears that the acquisition of TVA agricultural per-
sonnel from the extension service and the basic commitment of TVA to
carry on the program through the extension service seriously inhibited
the Authority's capacity to deal with the special problems of farm
tenancy.
While the relationship between TVA and the extension services natu-
rally implies that most influence should flow from the latter to the
former, the weight of TVA-subsidized activities and personnel has not
been unnoticed in the work at the county level. Work on the test-
demonstration farms, with county associations of farmers and commu-
nity organizations, provides a definite framework for the extension
program which otherwise often tends to be somewhat diffuse. There is,
therefore, a noticeable tendency for the center of work at the county
level to shift to the TVA assistant county agent. This has resulted in
some dissatisfaction among the men, who sometimes feel that they are
doing "the really important job in the county" and yet are held to the
status of assistant agents. This subordinate status is assured, however,
so long as the men assigned to the TVA program are placed within the
extension service chain of command. Moreover, the assistant agents are
not dependent upon county funds for their salaries, so that they tend
to be independent of courthouse politics. As a result, extension officials
find that some assistant agents are reluctant to take over the position of
county agent. Though the character of their work might well entitle
them to a higher rank and salary, the pay of a TVA-subsidized assistant
agent is deliberately kept below that of the county agent because, as
Selznick : T VA and the Grass Roots 141
one extension official said, "It's hard enough to get them to take over a
county as it is." The salaries of the assistant agents are determined by
the colleges, and it is the internal administrative problem of the exten-
sion service staff which may decide the amount paid, rather than TVA.
The TVA-subsidized personnel, including supervisors, specialists, and
assistant agents, sometimes exhibits a tendency to become a "little TVA"
within the college staff. This is a normal consequence of having a definite
set of activities and being paid out of special funds. Thus one supervisor
in the field, though officially unconnected with TVA, considered it ex-
pedient to know the TVA staff at Knoxville and especially to be ac-
quainted with the real center of authority within that staff. At times,
the college officials must make positive efforts to pull the staff together
within the single hierarchy. However, this tendency toward a split in
the extension service organizations is remarkably small, due in part
to the conscious effort made by the TVA agriculturists to maintain the
integrity of the extension service organization and halt all attempts to
by-pass the formal lines of authority. In 1942, the TVA held a training
conference for the assistant agents on the joint program, but some doubts
were expressed by the Director of Agricultural Kelations whether this
would not make the men think of themselves as TVA personnel, an
eventuality which would be contrary to the policy of both TVA and the
extension services.
The TVA agricultural program has been successfully integrated with
that of the extension service. This has been accomplished by beginning
with the assumption that (1) the TVA could not legitimately elaborate
a program which would be different from that advocated by the grass-
roots institutions of the area ; and (2) the already established extension
services represent the most logical and technically most adequate avenue
to the farm population and its problems.
KELATIONS TO THE FARM BUREAU
The traditional partnership of the extension service and elements of the
American Farm Bureau Federation 26 raises the question of the extent
to which a similar relationship has prevailed between the Farm Bureau
and the TVA. Theoretically, and in the light of the above discussion,
it might be expected that the TVA-extension service and extension
service-Farm Bureau relations would be transitive. On the basis of
available evidence, this appears true. The TVA agriculturists have ap-
parently simply carried over from their extension service days an out-
26 See pp. 119 ff. above.
142 University of California Publications in Culture and Society
look in which, close cooperation with this private organization and its
state and national leaders is taken for granted.
A contract little known even within T VA, entered into by the Author-
ity with the Tennessee extension service, is the basis of one phase of prac-
tical relations between the TVA and the Farm Bureau. This agreement,
known as Supplementary Project No. 5, under the master contract be-
tween TVA and the University of Tennessee, was signed in 1935 by
the Tennessee Director of Extension and the TVA Director of Agricul-
tural Relations and continued in effect from year to year without formal
renewal. A "statement of relationships" recalls that the Agricultural
Extension Service of the U. S. Department of Agriculture and the na-
tional and state Farm Bureau Federations have a history of cooperation
for mutual objectives; that the Tennessee extension service has under-
taken research and demonstration work on behalf of the TVA ; and that
the "advice, assistance and cooperation of the Tennessee Farm Bureau
Federation will be of great value in the conduct of such demonstra-
tions . . ." On the basis of this understanding it was stipulated :
I. That the Agricultural Extension Service shall arrange with the Tennessee Farm
Bureau Federation for the assignment of one or more of the principal officers thereof
who shall devote his or their full time in furtherance of the purposes of this project.
II. The Agricultural Extension Service of the University of Tennessee shall arrange
for the use of the experience, facilities and personnel of the Farm Bureau Federation
in the organization of conservation and terracing clubs, the distribution of fertilizer
and the conduct of such educational work as will result in a wider diffusion of the
mutual aims and objectives of sound agriculture through redirected farm practices
in a regional program of watershed protection. It is mutually understood and agreed
that all activities of the Tennessee Farm Bureau Federation pursuant to the terms
hereof shall be conducted without regard to membership in said organization.
III. The Tennessee Farm Bureau Federation shall interpret for the national or-
ganization of the American Farm Bureau Federation, and through it to the National
Council of Farm Organizations, tho aims and objectives of this cooperative program
of readjusted agriculture and watershed protection and shall from time to time indi-
cate to the Authority through the Extension Director the modifications in the plan
of procedure recommended by the National organization as being of special signifi-
cance in the solution of the national problem of land conservation.
This contract was also signed by the President of the Tennessee Farm
Bureau Federation as an agreement between that organization and the
University of Tennessee. A supplementary memorandum provided for
a special field agent, to be designated Director of the Educational Divi-
sion of the Tennessee Farm Bureau Federation and paid by the Farm
Bureau which would then be reimbursed by the extension service "from
money appropriated by the TVA for this purpose." Notwithstanding
Selznick : T VA and the Grass Roots 143
the provision of the last sentence of section II quoted above, it is reported
that the agent in question has functioned in fact as an organizer for the
Farm Bureau. One TVA official, a participant in the agricultural pro-
gram, said that he himself had attended a meeting at which the agent
spoke for the Farm Bureau, and that in general he felt the arrangement
to represent "an unhealthy subsidy of a private organization by federal
funds." "However," he continued, "this is not an unusual situation in
the relations between extension service and the Farm Bureau. Many of
the extension personnel do organization work for the Farm Bureau.
In Mississippi, the president of the state Farm Bureau is on the board of
trustees of the Mississippi State College and he exerts pressure on the
extension service to help in membership drives, which is why there are
about 30,000 Farm Bureau members in Mississippi, about 10 per cent
of the farmers." The report of this official with regard to the Farm
Bureau agent paid out of TVA funds has been confirmed by at least
one other individual.
Not only has the TVA continued the day-to-day relations to the local
Farm Bureau, which are normal in the extension service, but there is
also evidence that there has been close contact between leading agricul-
turists in the Authority and the national leadership of the American
Farm Bureau Federation. These relationships have extended to per-
sonal conferences on legislative matters and interagency controversies.
Thus, on June 27, 1941, Mr. J. F. Porter, President of the Tennessee
Farm Bureau Federation, wrote to Mr. E. A. O'Neal, President of the
American Farm Bureau Federation, in a vein which could leave no
doubt concerning the close relation between TVA and this national farm
organization. Mr. Porter reported that he had held a conference with
Dr. H. A. Morgan on the land-grant college situation. Dr. Morgan voiced
his concern regarding attitudes expressed in some quarters toward the
land-grant colleges and stated that he would like to cooperate with Mr.
O'Neal in the matter. He suggested that President Koosevelt should be
informed concerning the situation and that TVA might well intercede
if the large farm organizations would ask TVA to make a study of the
land-grant college controversy. TVA would then take up the matter with
the President. Mr. Porter indicated that this procedure would not take
much time, since TVA was already well informed. He also discussed the
question of TVA's jurisdiction over the Cumberland Valley area, indi-
cating that he had assured Dr. Morgan of the Farm Bureau's support.
Following these discussions, a meeting was held late in July, 1941 at
Muscle Shoals, Alabama, attended by Dr. H. A. Morgan, Mr. J. C.
144 University of California Publications in Culture and Society
McAmis, Mr. Neil Bass, Mr. Porter, and Mr. O'Neal, at which the impli-
cations of interagency controversy for the role of the TVA were dis-
cussed.
This evidence explicitly supports the interpretation that relations
with the Farm Bureau have been national as well as local and have in-
volved matters charged with political implication. It appears that Dr.
H. A. Morgan, in effect a representative of the agriculturists on the TVA
Board, has been ready to intercede with the President of the United
States on behalf of the land-grant college system. At the same time, the
President of the American Farm Bureau Federation, not ungrateful
for support of the institutional basis of his own organization, lends his
good offices in matters of legislative interest to the TVA. It will be re-
called (p. 78 above) that in 1941 the Department of Agriculture was
among those who opposed a bill for the inclusion of the Cumberland
Kiver basin within the jurisdiction of the Authority.
The close relation between the TVA and the Farm Bureau reinforces
the initial commitment to the extension service system. Alternatively,
it is evidence that the relationship to the extension service is something
more than a formal cooperative arrangement for stated objectives. The
concrete operation of the grass-roots policy has, in this respect, led to
consequences unlocked for by the founders of the Authority. One of the
most important of these founders was Senator George W. Norris, whose
views on TVA and the Farm Bureau are, consequently, of interest :
Senator NOBBIS : I did not intend to mention it, but since the Judge has brought it
out, I will say that I have had some correspondence with one of the farm organiza-
tions in criticism [of TVA's relation to the Farm Bureau]. I think it is proper for
me to say that, because I have been connected with the TVA, or with what the TVA
has been doing, and I was unaware that they were doing anything that I thought was
wrong about it. But I took the matter up ... with the idea of ironing out what I
thought was just a simple little . . . case of jealousy, that is probably natural, that is
almost human they thought the Farm Bureau was getting too much attention, and
they wanted some of it ... I tried to convince them, and I hope I succeeded, that there
was no possible ground for complaint that the TVA was picking out any farm organ-
ization ; that they were glad to cooperate with all sorts of organizations, and I called
their attention to the law, which provides what they should do, and that they had not
done anything with the Farm Bureau that they would not do with every other organi-
zation. 27
Senator Norris does not appear to have been fully informed in this mat-
ter. It was this same senator who, speaking of the Farm Bureau lobby in
1928, predicted ". . . the time will come when the rank and file of Ameri-
27 Hearings, Joint Committee to Investigate the Adequacy and Use of Phosphate
Eesources of the United States, 75th Cong., 3d sess. (1938), p. 202.
Selznick: T VA and the Grass Roots 145
can farmers will begin to realize by whom they are being deceived here
in Washington." 28
Apart from the general effects of the relationship with the extension
service, there is little indication of any special influence by the Farm
Bureau on the TVA program. However, it is reported that the Farm
Bureau in Tennessee has undermined efforts to organize farm coopera-
tives on the theory that a network of such cooperatives might become
sufficiently strong to be independent of the extension service and the
Farm Bureau. A proposal to organize an association of cooperatives is
said to have been quashed under Farm Bureau pressure. In general,
there is said to be passive resistance on the part of the Farm Bureau
and the extension service to the organization of cooperatives. Since the
TVA works only through the extension service in such matters, any
orientation of the Authority in that direction would be inhibited.
AN ADMINISTRATIVE CONSTITUENCY
The foregoing analysis of the relationship between the TVA and the
farm leadership in the Valley area may be summarized in the concept
of the administrative constituency. The idea of cooperation or working
with and through local institutions is qualified in its concrete reference
by the sociological dimension of administrative behavior. This dimension
recognizes the force of informal goals and institutional commitments,
and to that extent serves to make explicit some well understood but sel-
dom formulated bases of executive action. A constituency is a group,
formally outside a given organization, to which the latter (or an element
within it) has a special commitment. A relation of mutual dependence
develops, so that the agent organization must defend its constituency
and conversely. This relation gains strength and definition as precedents
are established in behavior and in doctrine, and especially as the constitu-
ency itself attains organized form. A group which finds its coherence in
common interest, but remains unorganized, may enforce its demands
in subtle ways, but a leadership and a machinery serve to mobilize its
resources. At the same time, however, this machinery may become sepa-
rated from its popular base and itself become the effective constituency.
The idea of an administrative constituency, however it may be
phrased, is familiar to students of public administration, 29 and in gen-
28 Hearings on Disposition of Muscle Shoals, Senate Committee on Agriculture and
Forestry, 70th Cong., 1st sess. (1928), p. 79.
29 See A very Leiserson, Administrative Regulation (Chicago: University of Chi-
cago Press, 1942), p. 9; also V. O. Key, Jr., "Politics and Administration," in L. D.
White (ed.), The Future of Government in the United States (Chicago: University
146 University of California Publications in Culture and Society
eral may be thought of as a normal mechanism of social control whereby
formal organizations are made responsive to relevant publics. Or, put
in other terms, the creation of the constituency relationship is a form of
cooptation, the informal involvement of local elements in the process
of policy determination. One TVA staff member felt that the TVA's
agricultural constituency operated as "one element in the general trend
in the Authority toward conservatism and adjustment to Valley insti-
tutions." It is this adjustment which is one of the significant implications
of the process of cooptation in administration.
The constituency relation varies widely in source, in intensity, and in
meaning. One form may arise simply out of the need felt by an organiza-
tion, public or private, to defend its continuing working relations with
an outside group or agency. Interest may focus only accidentally upon
the particular outside group involved, for it is the smooth avenue of
operation which is being defended, and will not be readily jeopardized
in the interests of a program in which the organization may have no
great stake. It is often inconvenient and difficult to alter established
procedures, so that a given form of cooperation may be defended in
order to preserve the integrity and equilibrium of operations. This situa-
tion is analytically quite different from that in which the outside organi-
zation is defended for its own sake, as occurs in the relation between the
TVA Agricultural Relations Department and the land-grant colleges.
There is also a distinction to be drawn between short-run pressures upon
an agency and the long-run strategy by which a measure of continuing
control is achieved. In connection with the administrative relations
analyzed above, the land-grant colleges have sought and gained a signifi-
cant measure of influence upon the TVA ; moreover, the TVA agricul-
turists do defend the colleges as valuable, and even indispensable, in
themselves. That a constituency relation in this advanced form exists
is well known in the higher circles of TVA, and is the subject of much
comment.
The significance of a constituency relation depends in part upon the
fact that the character of the constituency will tend to define and shape
of Chicago Press, 1942), p. 151: "In one respect individual parts of the administra-
tion tend to become like private pressure groups in that they have their own par-
ticularistic and parochial interests to defend and promote. . . . The association of
the agency with its clientele sometimes makes for a 'representative bureaucracy.'
The policy drives of the hierarchy arising from the immediate interests of its mem-
bers are reinforced and colored by the power and wishes of outside groups concerned
with the work of the agency. Through administrative determination of delegated
policy questions and through administrative influence on new policy, the desires of
private groups may be effectively projected into the governmental machinery. In
extreme circumstances something in the nature of a guild may be approached."
Selznick : T VA and the Grass Roots 147
the character of the agency. This may involve the recruitment of per-
sonnel from the ranks of the constituency and, in an extreme form, the
assumption by elements within the agency of a leadership status with
respect to the constituency. The representatives of the constituency
within the agency then come to define their role as one of leading a broad
struggle for the furtherance of the interests of the constituency, and
sometimes may be more conscious of those interests than members of
the outside group themselves. Where the outside group is not the formal
source of public policy, the constituency relation may remain more or
less covert, and its representatives may find it necessary to devise and
rely upon some doctrine or ideology to cover and defend the real rela-
tionships. An adequate comprehension of the full meaning of the grass-
roots policy does not appear to be possible apart from some such
principle as this.
In the relation of TVA to the land-grant colleges, the mechanics of
representation include attempts by the Agricultural Relations De-
partment of TVA to (1) channel all activities which may possibly be
interpreted as within their subject-matter field through the land-grant
colleges ; (2) make itself the sole point of contact by the Authority with
the institutions; (3) actively oppose all encroachments on the preroga-
tives of the colleges; and (4) further the policies of the colleges within
the Authority, as opportunity may arise.
1. The TVA agriculturists have tried to enlarge the scope of the activi-
ties that are carried on through the colleges, to include responsibilities
of other subject-matter departments within the Authority. In this re-
spect the approach of the agriculturists has been not toward local
agencies in general, but toward the extension services in particular, even
when that may mean making a choice against possible alternatives among
the state agencies. The history of the TVA Forestry Relations Depart-
ment points to a different attitude toward local institutions from that
utilized by the TVA agriculturists, and provides concrete evidence of
the attempt to swing the foresters into line with the organizational meth-
ods of the agricultural group.
The TVA forestry division was originally organized as a part of the
A. E. Morgan wing of the Authority, under the leadership of E. C. M.
Richards. An aggressive program of reforestation and watershed and
reservoir protection was conducted primarily on a direct-action basis,
using camps of the Civilian Conservation Corps assigned to the TVA
by the Forest Service, and not hesitating to deal directly with local
landowners in the distribution of tree seedlings. Work was undertaken
148 University of California Publications in Culture and Society
with various agencies, including the State Foresters and the U. S. Forest
Service, but on the whole the TVA forestry group felt that it had a
program of its own, which could not be effectively delegated. However,
a convergence of interest with the extension services was soon evident,
and this became the basis of a protracted controversy with the Agricul-
tural Relations Department of the Authority. This convergence referred
to the joint interest of TVA and the extension service in farm forestry
work and in education for forest-fire control, both deemed by the exten-
sion service to be within the province of the county agent organization.
The professional foresters under Richards, however, did not believe that
the extension service could adequately carry out the program. At a meet-
ing early in TVA history, which included H. A. Morgan, J. C. McAmis
(Director of Agricultural Relations), and Richards, the latter outlined
the basis on which he thought the forestry division might enter farm
forestry work. This would involve having the forestry division itself
hire and direct the activities of men who would do the work, though co-
operating closely with the county agents. The agriculturists, however,
asserted that such a procedure would encroach on the field of the exten-
sion service, and requested that Richards approve turning over TVA
funds to the extension services for farm forestry work, allowing them to
hire and direct the activities of the men. Richards refused to do this,
and the matter lay dormant for some time afterward.
The attitude of the TVA Chief Forester, and the intensity of the con-
troversy, may be judged from the forceful statement made by Richards
in 1938, when the conflict within the TVA Board was under public inves-
tigation. This statement is especially important in the light of the entire
analysis thus far presented :
There are decided differences of opinion within the TVA as to just what the TVA is
or should be. Three views exist and fight persistently for acceptance :
1. That TVA is a federally owned and operated electric power utility corporation.
All else is quite secondary and is only permitted to exist because (a) it, like dam
construction, must go on if electric power is to be generated; or (b) the other activi-
ties are so generally popular and worthwhile that it takes more courage on the part
of the advocates of this "power concept" openly to throw it overboard, than to pay
lip service to it and quietly work under cover to hamstring it and keep it insignificant.
2. That the TVA is an adjunct to local agencies, such as the land-grant colleges
and extension services of the seven valley states. Except for dam construction and
similar work, which manifestly is quite impossible for the local and state agencies to
perform, the advocates of this concept believe that the Authority itself has nothing
in the way of a job to do ; that it should merely get money from Congress, pass it out
to local and state agencies, and let them do everything in their own way, with per-
sonnel of their own choosing. Knowing little or nothing of either forestry or wildlife,
Selznick: TVA and the Grass Roots 149
this group holds firmly to early pioneer ideas of land ownership, land use, and con-
servation. It thinks of forests as small scattered farm woodland tracts. It thinks of
foresters as extension workers to be called in for advice and suggestions by the county
agricultural agents and farm owners. As the farm woods is, in their minds, only a
minor part of the farm, so the forester or the wildlife expert is only a minor tech-
nician, quite unfit for large responsibility in land management and policy determina-
tion. One can imagine the attitude of this group toward the many important problems
of forestry and wildlife management confronting the TVA.
3. That the TVA is a unique effort on the part of the people of the United States,
through the federal government, to solve the major problems of a great watershed.
The advocates of this third concept propose to approach the job with a scientific
attitude. They have been assembled from all over the U.S. ; they are spending most
of the money and are doing most of the work. To them the TVA is simply a large
federal organization, with a very clearcut, definite job. They believe that the other
federal and state organizations are important, and that the TVA should cooperate
with them fully, but that TVA should not turn over to them, however, the work of the
TVA itself. 30
An examination of the record indicates that the TVA forestry division
has not had an exclusive orientation toward any given set of local institu-
tions. Thus, while lacking confidence in the extension service as a whole
as far as forestry work is concerned, the division has made efforts to
strengthen the hand of the extension forester within his own organiza-
tion. There has also been cooperation with the vocational agriculture
teachers, and good relations with the Forest Service. The dominant note
seems to be one of promoting the professional interests of forestry, wher-
ever this may lead. To some extent, this has in practice involved an effort
to build up the state conservation departments and forestry divisions,
but on the whole these have not been sufficiently strong to form a firm
constituency. Moreover, the TVA foresters would tend to avoid being
tied too closely to the conservation departments, for that might result
in exclusion from work on farm woodlands. In general, it may be said
that the TVA forestry division has followed perspective 3 listed above
in the quotation from Richards.
With the victory of the Lilienthal-H. A. Morgan bloc in the TVA as a
whole, and the concomitant departure of Richards, pressure for the re-
orientation of the forestry division along lines desired by the agricul-
turists was renewed. Despite the change in leadership, the forestry
division retained its direct-action viewpoint, though in a somewhat
chastened form. In the general reorganization of the Authority, the
Forestry Relations Department was brought under the supervision of
a Chief Conservation Engineer, whose office functioned as a convenient
80 E. C. M. Richards, "The Future of TVA Forestry," Journal of Forestry, 36 (July,
1938), 644-645.
150 University of California Publications in Culture and Society
mechanism for the exercise of control over the foresters to implement this
reorientation. In 1937, on the initiative of the TVA agriculturists, the
colleges were requested by the Authority to recommend a forestry policy.
Not unnaturally, the institutions recommended that TVA follow the
same procedures in the farm forestry program as were utilized in the
agricultural work. It is significant for understanding the meaning of
such formal mechanisms that the Agricultural Relations Department
was able to use (1) this formal recommendation from the colleges, (2)
the original three-way memorandum of understanding, and (3) the
precedents established in its own activities, as arguments to buttress
the attempt to bring the TVA foresters into line with its own procedures.
The work of the Forestry Relations Department with CCC camps was
discontinued in 1942. This, in addition to other wartime developments,
provided the occasion for the agriculturists to renew the attempt to alter
the orientation of the foresters. After conferences with representatives
of the Forestry Relations and Agricultural Relations Departments, the
Chief Conservation Engineer "laid down the line," insisting that the
foresters carry out the obligations of the Authority so as to strengthen
the land-grant colleges "in the performance of their responsibilities to
the people of their states and so as to avoid duplicating and confusing
field administration." Specifically, he said that this was to be done by
adding additional personnel to the extension service staffs. It is safe
to say that the foresters were less than enthusiastic about this injunction,
and by 1943, a total of only about five men were either assigned to the
administrative jurisdiction of the extension services or hired by the lat-
ter with salaries reimbursed by TVA. Moreover, the contract with the
University of Tennessee for the employment of two foresters (assistant
county agents) indicated that the Forestry Relations Department was
reserving more control over the administration of its contracts than was
customary in the Agricultural Relations Department. This contract in-
cluded provisions for TVA approval of the assistant agents selected and
for inspection by TVA of their technical performance.
In general, even after the departure of Richards and others of the
A. E. Morgan group, the TVA foresters remained unconvinced of the
value of the procedure advocated by the agriculturists. This is a matter
of method in relation not only to the extension services, but to all local
agencies. Thus the foresters, in supporting and strengthening the state
conservation departments, do not favor grants-in-aid or subsidies by
TVA. They follow rather the way of building public support for appro-
priations to these agencies on a county and state level, as well as for
Selznick: TVA and the Grass Roots 151
further funds from the federal government under the provisions of the
Clarke-McNary Act. Favorable public sentiment is encouraged through
a forest-fire educational program carried on at the county level, and by
official TVA representations to the President as to the states' need for
increased funds for work in forestry. This general orientation, added
to the initial lack of faith in the ability of extension service to carry on
a forestry program, makes them loathe to turn over personnel to that
organization. But because of the prestige of the grass-roots method
within the Authority and the power of the agriculturists, the TVA for-
esters seem to be fighting a losing battle.
2. In exercising its function as the representative of a constituency,
it is convenient for the TVA agriculturists to make themselves the pri-
mary or even exclusive channel of the Authority to the colleges, which
insures that direct contact will be made through individuals who will
view the joint activities with friendly eyes and frame their reports in
terms which will uphold the established relationships. Two examples
may suffice. In 1941, the TVA Director of Agricultural Relations at-
tempted to restrain representatives of the TVA Comptroller from
examining field reports of the extension service at the institutions, in
connection with audits of reimbursements under cooperative contracts ;
he insisted upon a literal interpretation of a clause in the project agree-
ment which states that "during the progress of the work the Chief of
the Agricultural Division of the TVA, or his representative, shall have
access to any records or reports pertaining to the progress of the work
hereunder." Thus the agreement so framed is a prop of the constituency
relation, protecting the colleges against possible unfriendly attacks.
Both the colleges and the TVA agriculturists interpret their agreement
as not with the TVA as a whole, but with the Department of Agricultural
Relations. This tactic may also be useful to the TVA agriculturists in
holding the institutions in line, lest members of the college staffs be
overly cooperative with other branches of TVA and thus serve to loosen
the established ties. The restriction of contact serves not only to protect
the colleges but also to preserve the existing constituency relation.
To this more or less routine effort we may add as a further example
the attempt of the TVA agriculturists to formally establish themselves
as the coordinators of all activities by the Authority which affect agri-
culture in any way. This is the proposition, mentioned above (p. 116),
for the promulgation by central management of an Administrative Code
which would solidify the position of the land-grant colleges and give the
Agricultural Relations Department authority to approve or disapprove
152 University of California Publications in Culture and Society
all proposals initiated by other departments in the TVA and affecting
agriculture or rural people. With the Authority operating in a predomi-
nantly rural area, the significance of such control exercised by a group
in TVA with special commitments to outside elements would be very
great. Up to June, 1943, no action had been taken by central manage-
ment on this proposal. If adopted, the code would give the Agricultural
Relations Department a measure of control over the activities of a
number of TVA departments whose activities affect rural people, in-
cluding Commerce (agricultural engineering developments), Forestry
Relations, Regional Studies, Power Utilization (rural electrification),
Health and Safety (malaria control work in rural areas) , and Reservoir
Property Management (contact with farmers in the administration of
TVA-owned lands) .
3. The TVA agriculturists have, during the history of the Authority,
found a number of opportunities to defend the extension service organi-
zation and outlook from encroachments by other departments in the
Authority. This was more evident in the early days of the TVA, when
efforts to think in terms of regional planning in its broader sense were
more common. The agriculturists were known as the bitter enemies of
the Land Planning Division, composed of what they called, pejoratively,
"the geographers." Reports issued by other departments affecting rural
areas were subjected to severe criticism by the agriculturists, and in
one case at least, as late as 1939, an attempt was made to ban the distribu-
tion of a study of rural zoning which had been carried out not by the
TVA alone but cooperatively with the Bureau of Agricultural Eco-
nomics of the U. S. Department of Agriculture, the Tennessee State
Planning Commission, and the Hamilton County (Tennessee) Regional
Planning Commission. The TVA's Departments of Regional Planning
Studies and Forestry Relations of TVA had participated. In the same
year, another report compiled by the Regional Studies Department on
the probable effects of one of the TVA projects upon economic and social
life in the area was criticized by the Director of Agricultural Relations :
1. There is little, if any recognition of the prior right and responsibility of people
in local areas to take part in the planning and execution of the project under con-
sideration.
2. The report does not capitalize on the judgment and experience which resides
only in the local areas and can be made available only through organized study and
analysis by the local people.
3. The conduct of investigations into the private affairs of individuals without
their collective concurrence goes beyond what is proper for a Federal agency and
abridges and undermines the relationships which have been accepted as policy by the
local, state, and Federal agencies.
Selznick : T VA and the Grass Roots 153
4. Without the collective judgment of responsible organized groups, it is doubtful
if information obtained is reliable and trustworthy as a basis of action of such far-
reaching consequences in local areas.
I reiterate my attitude, many times expressed on similar matters, that studies made
and conclusions drawn with reference to agricultural situations and problems in local
areas should be done by, and largely for, the people themselves, acting through their
local, state, and Federal educational agencies.
The implication of these remarks clearly seems to be that only such insti-
tutions as the land-grant colleges may rightfully prepare reports on
matters affecting farm people. It is doubtful that the Director of Agri-
cultural Relations would consider a local planning commission, organ-
ized with the aid of some other TVA department, a bona fide "local
institution.'*
4. Examples of the furtherance of specific policies of the land-grant
colleges by the TVA agriculturists will be detailed in chapters v and vi
below, indicating that the role and character of the TVA as a whole have
been shaped, at least in part, by the cooptative relationship in the field
of agriculture.
In conclusion, we may consider the mutual dependence of the Agri-
cultural Relations Department as agent, and the land-grant colleges as
constituency. Two views are presented by persons close to the program.
One suggests that the colleges "don't really give a damn" about TVA or
even the money which it turns over to them that the colleges actually
have a whip hand over the Agricultural Relations Department in the
possibility of their withdrawal from the joint program. Since the depart-
ment is so deeply committed to the theory that it is impossible to carry
on a program apart from the colleges, such an eventuality would be
disastrous. On the other hand, it has been suggested by a leading par-
ticipant that Director McAmis "has strong control over the state exten-
sion services, who fear a possible withdrawal of TVA funds." Doubtless
both the constituency and its agent are mutually dependent upon each
other, but it seems probable that the dependence of the TVA agricul-
turists on the preservation of the established relation is the greater. A
reorganization of TVA agricultural activities might change the form of
cooperation with the extension services and vitiate some of the conse-
quences of the earlier relationship, but cooperation in some form would
continue. A change in policy by the top TVA leadership might result
in the removal of the existing group in control of the agricultural pro-
gram though with some sacrifice of TVA's position in the Valley, but
it could not seriously affect the administrative existence of the extension
service.
CHAPTER V
UNANTICIPATED CONSEQUENCES, 1: THE
STRUGGLE IN AGRICULTURE AND
THE ROLE OF TVA
I'm not opposed to security spelled with a small 's' ; but when you use a large l s,' it
leaves me cold. DIRECTOR OF AGRICULTURAL EELATIONS, TVA ( 1943 )
THE EXERCISE of administrative discretion gives life and meaning to the
abstractions of policy and doctrine. 1 The injunctions of a statute, the
directives of a central management mobilize action on the assumption
that common organizational goals exist and that these goals are effective
in shaping rational administrative behavior. But this assumption breaks
down as, in the exercise of discretion, officials are faced with the need
to deal selectively with their environments. For as they come upon an
existing social situation, individual administrators find it difficult to
restrict their interest and involvement to the formal goals of the policy
they are executing ; they tend to be involved wholly, bringing to bear
their own fractional interests upon day-to-day decisions. These special
commitments may be well organized and consistent, or diffuse and spo-
radic. They range from the extreme case, in which the whole allegiance
of the official is to some outside group or interest, to the relatively mild
and thoroughly normal attention which a field representative may give
to bolstering his position within the organization. The former case may
require drastic correctional measures, whereas the latter may involve
only minor changes in the mechanics of central control. But whatever
the forms, or the intensity of their expression, the centrifugal tendencies
inherent in the delegation of discretionary powers must be recognized
and taken into account by those responsible for the integrity of the whole
organization.
In the analysis of this centrifugal tendency, we are led on theoretical
grounds to the following considerations: (1) The organization may
enter upon a preexisting social situation in which there are loci of power
which will not permit themselves to be ignored as activities in which
they have an interest are initiated. (2) Intervention may occur in a
context of controversy among forces within the area of operation, so
that the problem of influencing the controversy may be posed. (3) The
intervening organization may be actually or potentially split into f ac-
1 See above, pp. 64-68.
[155]
156 University of California Publications in Culture and Society
tions, thus enhancing the possibility that one group within the organiza-
tion may exercise whatever discretion may be delegated to it in terms
of tangential goals and commitments. (4) As the discretion of a part
of the administrative apparatus is exercised in the name of the organ-
ization as a whole, the latter may find itself committed to a policy or a
set of relationships which it neither anticipated nor desired. Thus the
parochial interests of a subordinate part of an organization must be
viewed in the light of the social situation within which operations are
carried on and in terms of long-run consequences for the position of the
organization as a whole.
If we combine these theoretical considerations with the preceding dis-
cussion of the TVA's agricultural machinery, it will be possible to draw
out more fully the meaning of the grass-roots idea as it has worked out
in practice. This meaning is explicated as consequences are traced and
the practical force of the doctrine observed. The consequences are ob-
jective, in the sense that they have been determined by forces generated
willy-nilly from the initial relationships established; they are unin-
tended in the sense that they are not the predicted products of executive
action. The grass-roots policy suggests that the program of a regional
agency may be channeled through existing agencies, but does not, on
the level of doctrine, take into account the possible tangential results
of such a procedure.
There are, as we have noted, three levels which define the "decentral-
ized administration of federal functions." These are (1) the location of
an agency in the area of operation, with sufficient managerial autonomy
to enable it to deal effectively with local conditions in local terms; (2)
the method of working with and through local agencies in the execution
of a program; (3) the encouragement of participation by the people
themselves in administration, primarily through the use of voluntary
associations. The first of these conditions raises the problem most di-
rectly of the relation of such an agency as TVA to the rest of the federal
structure. The second condition seems to be independent of the first,
concerned as it is with the relation of the agency to the local institutions.
But this mutual exclusion is only apparent. As selection and interven-
tion with respect to the local area occur, it becomes clear that many local
decisions will be made in terms of their national consequences. Local
intervention may offer an opportunity to influence a far-reaching de-
cision which at first sight is irrelevant to the specific discretionary
problem posed. 2 Consequently, it may be anticipated that the concrete
2 See above, p. 73.
Selznick: TVA and the Grass Roots 157
exercise of discretion, in relation to the principle of working with and
through existing institutions in the area, may seriously affect the rela-
tion of the agency as a whole to the federal system. This inference is
borne out by the TVA experience, as will be made clear in the course
of this chapter.
INTERAGENCY RIVALRIES IN AGRICULTURE
The initial organization of the TVA in 1933 was immediately faced with
the existence of the land-grant college system in the field of agriculture.
As we have seen, 3 the potential issue was resolved in advance by the
absorption of the TVA agricultural program within that of the exten-
sion service and, more specifically, by the establishment of the constitu-
ency relationship between the colleges and the Agricultural Relations
Department of TVA. But the Authority entered upon a social situation
which not only contained preexisting organizations having established
prerogatives but was, in common with the rest of the nation, the scene
of a protracted controversy among contending social forces for control
over the agricultural machinery of the United States government. In
this controversy, the TVA as a whole, borne along by the constituency
relation already established, became committed to one of the contending
sides.
For almost a generation after the establishment of the agricultural
extension service in 1914, the contact of agencies of the government with
the farm population remained relatively simple. The county agent sys-
tem represented the main channel through which information and, from
time to time, exhortation was brought to the farmer/ Twenty years of
operation, added to the earlier years of preparation, served to stamp the
extension service organization and personnel with a traditional charac-
ter, inevitably shaping its conception of the problems with which it
might be faced. 5 The stress on education made for rather easy-going
operations, aiming at a gradual improvement of farming methods rather
than any immediate ameliorative measures. Knowledge, rather than
social or economic organization, was considered the limiting factor : the
objective of "growing two blades of grass where one grew before" was
uppermost. The spread of knowledge by example and demonstration
8 See above, chaps, iii, iv.
*See Russell Lord's informal history, The Agrarian Revival (New York: Ameri-
can Association for Adult Education, 1939). Also Gladys Baker, The County Agent
(Chicago: University of Chicago Press, 1939). Earlier, and more detailed, is the two-
volume Survey of Land-Grant Colleges and Universities, Office of Education, U. S.
Department of the Interior (Washington: Government Printing Office, 1930).
5 See above, pp. 117-124.
158 University of California Publications in Culture and Society
smoothed the way to the substantial farmer of the county who could
readily participate in new ventures and gradually set the pace for his
neighbors. In this way, the county agent system developed a set of pre-
dispositions and habits of work which clashed with the demands of those
who later called for aggressive action. Moreover, the association of the
agents with the more prosperous farmers tended to compromise their
position as class forces within agriculture began to erupt. 8
The break in prices and land values which followed the heady expan-
sion during World War I resulted in a persistent agitation for govern-
ment aid to the farmers. This pressure found its greatest response after
the advent of the New Deal administration, which opened a new era in
agricultural policy. A combination of measures for relief, production
control, soil conservation, and vastly expanded credit facilities served
to pose a series of new and pressing problems for those involved in the
network of administrative and political organizations in agriculture.
These new problems crystallized around two basic issues : (1) the emer-
gence of internal conflict among the farmers, according to distinctions
of economic class, and reflected in the alignments of farm organizations ;
(2) the change in the character of federal intervention in agriculture,
reflected in a challenge to the older governmental institutions, as new
agencies came upon the scene. These two issues were merged in the actual
struggle, in which agencies functioned as weapons in class conflict.
The American Farm Bureau Federation was a leading contender in
the agricultural struggle during the decade following 1933, and the
issues in which this organization was involved most clearly defined the
lines of controversy. The special role of the Farm Bureau was deter-
mined by its identification with the well-to-do and "industrial" farmers
as a class, and with the extension service system as its organizational
weapon. The class attitude was disclosed most clearly in the opposition
of the Farm Bureau to the Farm Security Administration. The latter
directed its program toward the underprivileged groups in agriculture,
especially the tenant farmers, and was supported by the Farmers Edu-
cational and Co-operative Union of America as well as by such smaller
groups as the Southern Tenant Farmers Union and the Farm Holiday
Association. The Farm Bureau bitterly attacked the entire program
of the FSA and demanded that its functions be either scrapped or
turned over to other agencies. 7 Some light on the class basis of this attack
6 See above, p. 121.
7 The Farm Bureau contented itself with ignoring the FSA in its appropriation
recommendations before 1940. Later, especially in 1942 and 1943 under the impetus
of an economy drive in Congress, it conducted an all-out attack. The conclusion seems
Selznick : T VA and the Grass Roots 159
is cast by the following statement attributed to a high official of the FSA :
Two fights over income are going on in American Agriculture. One has to do with
Agriculture's getting its fair share of the total income of the nation the "parity"
idea that the Farm Bureau has championed for years. The AAA is the key agency
in this, which explains why it has been so long and bitterly fought by many persons
outside the ranks of Agriculture.
The other fight is inside Agriculture's own family over distribution within the
family of whatever income Agriculture gets. FSA figures sharply in this, as champion
of the group that has long been on the small end. It follows logically that the bit-
terest enemy of FSA in this fight would not be a disinterested group outside Agri-
culture's fold, but the established, well-to-do group within the family that has been
enjoying the biggest share in the past and the Farm Bureau Federation tends to
represent that group. 8
In more specific terms, another supporter of the FSA stated :
The real reason for the bitterness lies in the fact that certain types of agriculture
the cotton plantation, the truck farm, the orchard require a large and easily avail-
able supply of cheap labor. This labor is drawn mostly from the poorest, the wretch-
edest, and the most shiftless of all our farm people. These are also the people whom
the FSA undertook to "rehabilitate." The of tener the FSA succeeds and they have
been at it for nearly ten years the more the commercial farmer's labor supply
diminishes. And, to the extent that the "rehabilitated" become more than subsistence
farmers, the more competition there is for the market. 9
The latter interpretation is supported by a reading of testimony at con-
gressional hearings on the FSA. Thus, in March, 1943, the National
Cotton Council, of which the American Farm Bureau Federation is an
important constituent, attacked the FSA on the grounds that
. . . this agency is so functioning and so conducting its activities as to promote gross
inefficiency in the matter of culture and production of cotton and cottonseed; to
seriously impede cost of production ... to lower the morale of farm workers ... to
threaten, disturb, and disrupt economic and social conditions and relationships
throughout the Cotton Belt; to promote class distinctions, hatred, prejudice and
distrust within the Cotton Belt ; to threaten those who produce cotton and cottonseed
on a commercial basis ; to depress morale of cotton farmers throughout the belt, and
ultimately to destroy the business of farming as a free enterprise and a respectable
means of earning social and economic security by American farmers. 10
inescapable that, had the TVA undertaken an agricultural program along the lines
of the FSA, the formidable opposition of the Farm Bureau would have had to be
faced and, conceivably, might have turned the scale in favor of those who wished
to dismember the Authority.
8 Quoted by Carlyle Hodgkin, "Agriculture's Got Troubles," the second of a series
in Successful Farming (April, 1943), p. 26.
9 Oren Stephens, "FSA Fights for Its Life," Harper's (April, 1943), p. 478.
10 Oscar Johnston, President, National Cotton Council of America. Hearings, House
Committee on Appropriations, Agriculture Department Appropriation Bill for 1944,
78th Cong., 1st sess. (1943), p. 1619.
160 University of California Publications in Culture and Society
On another occasion, the Farm Bureau representative introduced as
evidence against the FSA a letter from a landowner protesting that the
FSA was "taking away tenants from landowners." 11 Again, President
Edward A. O'Neal of the Farm Bureau registered his support of pro-
tests against the FSA practice of including poll tax obligations among
allowable expenditures of tenant farmers in Alabama. 12 These considera-
tions, among others, indicate that opposition to the FSA has been
programmatic and substantive rather than merely organizational or
administrative. 18
The organizational phase of the controversy has, however, been equally
significant. As a mass organization, the Farm Bureau has been highly
conscious of the need to protect its avenue of access to the farm popula-
tion. To the extent that the Farm Bureau is identified with whatever
benefits farmers receive from governmental programs, its chances for a
large membership, and for control over farmer sentiment, are vastly
enhanced. This opportunity was for many years upheld by the special
relation of the Farm Bureau to the extension service organizations. 14
The president of the Iowa Farm Bureau Federation is reported to have
stated that "without the extension service the Farm Bureau would die." 15
The Farm Bureau is the consistent champion of the extension service
in the protection and enlargement of its appropriations and in defense
of it against competing agencies. The county agent system functions to
build and preserve the Farm Bureau membership. In defending the
extension service, the Farm Bureau fulfills an elementary condition of
its own survival.
The Farm Bureau's defense of the extension service may, however,
assume proportions which go beyond the needs of the college organiza-
tions themselves. The land-grant colleges are well established, and in
many areas need not fear the existence of other agricultural agencies
11 Donald Kirkpatrick, General Counsel of American Farm Bureau Federation.
Hearings, Joint Committee on Eeduction of Nonessential Federal Expenditures, 77th
Cong., 2d sess. (1942), p. 815.
12 Ibid., p. 742.
13 See J. M. Gaus and L. O. Wolcott, Public Administration and the U. S. Depart-
ment of Agriculture (Chicago: Public Administration Service, 1940), p. 195: "The
Administration of the [Agricultural Adjustment] program also had to meet the
pressures of commodity groups, the organization and support of which had long sus-
tained a demand for national farm program. The Administration had to cope with
the pressures of groups that, since they represented primarily the owner class, pressed
for policies favorable to themselves regardless of the interests of tenants and share-
croppers. In making these demands effective, the AAA came into conflict with a
froup in the Department that, to compensate for the injustices, became a motivating
actor in the development of the F.S.A."
14 See above, pp. 119-121.
13 Editorial, National Farm Holiday News (January 1, 1937).
Selznick : T VA and the Grass Roots 161
so long as their own integrity as educational institutions is preserved.
Other agencies, indeed, may even overshadow the colleges in importance
without producing panic among the latter, since the colleges continue
to have an enduring function. But for the Farm Bureau the extension
service represents its avenue of access to the farm population and, to
the extent that that access is compromised or dwindles in importance,
the Farm Bureau leaders may expect that farmers will cease to turn to
them for leadership. The Farm Bureau is therefore driven to push the
extension service into accepting ever increasing responsibilities. The
problems posed as governmental agencies come to function (or threaten
to function) as quasi-political machines have involved the Farm Bureau
and the land-grant colleges in a series of persistent difficulties.
The Farm Bureau's role as a representative of agricultural interests
in Washington has moved it to sponsor new modes of intervention by
the federal government for the benefit of the farmers. Perhaps the
farthest reaching of these in recent years has been the Agricultural
Adjustment Administration. The Farm Bureau "fathered" the AAA
Act of 1933 16 and was one of the chief supporters of the New Deal agri-
cultural program in its early and major phases. But its dilemma con-
sisted in this : in promoting the new program, it could not avoid the
construction of a new organization, the AAA. The latter, growing strong,
tended to have influence upon the farmers in its own right, and thus to
vitiate the prestige and strength of the Farm Bureau. This problem was
foreseen, for since the early days the Farm Bureau insisted that ad-
ministration of the program be channeled through the extension service.
Much was done toward that end, 17 but the large amount of patently non-
educational work involved and a growing disaffection in Washington
with the loose Smith-Lever relation of the Department of Agriculture
to the states 18 resulted in the construction of an independent AAA
organization. This included a network of county and community com-
mitteemen participating in the administration of the program. These
committeemen numbered approximately 90,000 and were paid about
$7,500,000 out of AAA funds in 1941. 19 As the years went by, the attitude
of the Farm Bureau toward the AAA cooled markedly, and by 1943 the
10 See Clifford V. Gregory, "The American Farm Bureau Federation and the AAA,"
Annals of the American Academy of Political and Social Science, 179 (May, 1935),
152-156.
17 See Baker, op. cit., pp. 70 ff., for details of aid derived by the extension service
from AAA funds.
1S See Lord, op. cit., p. 163.
19 See testimony of E. A. O'Neal, Hearings, Joint Committee on Reduction of
Nonessential Federal Expenditures, 77th Cong., 2d sess. (1942), pp. 754-755.
162 University of California Publications in Culture and Society
undercurrent of antagonism broke out in open conflict. 20 On April 1,
1943, the Tennessee Farm Bureau News carried the following headlines :
TRIPLE A LEADERS GUNNING
FOR AMERICAN FARM BUREAU
MOUNTING EVIDENCE REVEALS
OUT TO DESTROY
ORGANIZATION
O'NEAL SAYS
It was charged that AAA officials were attempting to antagonize farmers
toward the Farm Bureau because of the latter's stand against the con-
tinuation of AAA benefit payments. While the occasion for the outburst 21
was the stand of the Farm Bureau on this particular issue, the funda-
mental problem was that of organizational rivalry.
The Farm Bureau's answer to its dilemma with respect to AAA was
to insist that all educational, informational, and advisory services needed
in connection with the AAA program in the states, counties, and com-
munities be turned over to the extension service. This would insure that
the agency which dealt with the farmers would be friendly to the Farm
Bureau and help it to retain its mass support. This solution was pro-
posed consistently for all the new agricultural agencies which threatened
to construct an organization reaching down to the farm population and
which the Farm Bureau could not expect to control. Thus the Farm
Security Administration represented not only a class antagonist but an
organizational rival. The FSA did not carry on its work through the
extension services; it built its own field organization, with FSA repre-
sentatives at the county level dealing directly with farmers. This pre-
sented the dangerous possibility that the county agent's office might
cease to be the primary means of access to the farmer a possibility
inherently inimical to the interests of the Farm Bureau. The Bureau
consequently demanded that FSA be dismembered, with its farm and
home management services transferred to the extension service.
20 The underlying hostilities were of course well understood in agricultural circles.
One reporter wrote in March, 1943 : "In the beginning the American Farm Bureau
Federation was the great champion of AAA. It fought twenty years for the prin-
ciple of 'parity price.' It battled valiantly for the farm program in the halls of
Congress. It steered the new program through its early wobbly days. But now that
AAA has grown big and powerful, with a committee of five farmers in every precinct
in the U.S., the Farm Bureau begins to fear it as a rival farm organization." Carlyle
Hodgkin, Successful Farming (March, 1943), p. 54.
21 This issue of the Tennessee Farm Bureau News published lengthy quotations
from speeches of AAA officials attacking the Farm Bureau extension service bloc.
Selznick : T VA and the Grass Roots 163
The Farm Bureau also saw a threat to its position in the growth of
the Soil Conservation Service, another organization having its own men
in the field with direct access to the farmers. The program of the Farm
Bureau with respect to SCS, though lacking the class antagonism which
marked its attitude toward FSA, still demanded that the land-grant
colleges be utilized, rather than the SCS. Under its program,
. . . the work of the Soil Conservation Service in Washington would be handled by
the appropriate agencies which were established within the Department of Agricul-
ture to do such work ; regional offices would be abolished ; research and experimental
work in soil conservation would be transferred to the State Agricultural Experiment
Stations and carried on in cooperation with the U. S. Department of Agriculture;
educational, demonstrational, and advisory assistance to farmers and Soil Conserva-
tion Districts would be transferred to the State Extension Services and carried on
in cooperation with the Department of Agriculture. 22
In this way, the Farm Bureau consistently defended its most vital
asset its nearly exclusive access to the farm population through the
machinery of the agricultural extension service.
The dilemma of the land-grant colleges had the same basis as that
which plagued the Farm Bureau, but the alternatives were different.
The challenge of the new agencies seemed to demand that the colleges
offer their own organizations for the administration of the action pro-
grams; and this impulse was heightened by the desire of the Farm
Bureau that they do so. At the same time, the colleges desired to main-
tain their integrity as educational institutions, leaving the task of
administering a planned agricultural program to others. Many of the
colleges desired to avoid being identified with an agricultural program
which might be too closely linked to the current administration in Wash-
ington, or which might force the colleges to be the instruments through
which penalties against their constituent farm populations might be
enforced. Nor did the colleges respond uniformly. In the Southern
states, the weakness and relative poverty of the institutions made the
new program especially welcome. But in some of the more wealthy farm
states, as in Iowa, 23 there was a stronger tendency to eschew administra-
tive responsibility for the action programs.
Despite a certain ambivalence in the attitude of the colleges, it was
clear that on the whole they had little sympathy for those newer trends
in the Department of Agriculture which reflected a basic change in
22 The Nation's Agriculture (organ of the American Farm Bureau Federation)
(March, 1942), p. 2.
23 See The Sole of the Land-Grant College in Governmental Agricultural Programs,
Iowa State College Bulletin (Ames, la.: June, 1938).
164 University of California Publications in Culture and Society
attitude toward the methods by which agricultural programs were to
be administered. Many of the newer officials in the Department had little
confidence in the extension service 24 and presumably the more extreme
elements served to create a favorable climate of opinion for a revamping
of the official procedure. Milton S. Eisenhower said of this :
Most of the programs involved in this new comprehensive policy required planning
and action by farm people. They required also cooperative action by Federal, State,
and local agencies. But the cooperative action was of a different kind. It departed
from the traditional relationship of the grant-in-aid type that had existed between
the Colleges and the Department in research and extension work. In framing the
new laws the Congress, seeking to bring about simultaneous nation-wide action,
placed on the Secretary of Agriculture the responsibility for their administration;
the new laws did not authorize him to transfer his responsibility to any other person
or agency. Naturally, the laws authorized cooperation. 25
Thus it was not only the problem of facing competition in the field, but
also that of a change in the traditional and secure policy of the depart-
ment in the administration of its functions, which disturbed the land-
grant college chiefs. This issue created an atmosphere of controversy
in the field as well as what has been referred to as a split within the
Department of Agriculture itself in Washington. 26 To be sure, the more
dramatic aspects of this controversy within the Department may lead
to overfacile conclusions as to basic divisions in function. The "new"
Department is a product of the "old," and the action programs are the
result of a long evolution within the Department. 27 However, the spirit
of controversy, though generated by the relatively passing and tempo-
rary needs of organizational prestige and survival, is a significant factor
which must be taken into account in analyzing the behavior of interested
individuals and groups as they exercise administrative discretion.
TVA AND THE FARM SECURITY ADMINISTRATION
The leadership of the TVA's agricultural program has been an institu-
tional leadership. As such, it has conceived of its own role as larger than
the administration of a fertilizer program, justifying a self-conscious
24 See Lord, op. cit., p. 162.
23 Milton S. Eisenhower, "Who Should Be Eesponsible in the Development of an
Agricultural Planning Program?" an address by the Land-Use Coordinator and Di-
rector of Information, U. S. Department of Agriculture, at the meeting of the Asso-
ciation of Land-Grant Colleges and Universities, Chicago, November 14, 1938 (mini.).
26 See Stephens, op. cit., p. 481 : "There arose consequently a conflict between two
agricultural groups, and indeed, between two groups within the USDA. To be specific,
this conflict was between the Extension Service, the AAA and the Farm Bureau on
one side, and the FSA and the Farmers Union on the other. Basically, this is the
line-up as it exists today [1943]."
27 See Gaus and Wolcott, op. cit., p. 66.
Selznick: T VA and the Grass Roots 165
approach to the evolution of the pattern of organizational control in
agriculture. Keenly aware of its function as representative within the
TVA of the land-grant college system, the Agricultural Relations De-
partment has scrutinized all cooperative arrangements affecting farm
organizations and rural areas with an eye to the possible consequences
for its constituency. 28 In defending the colleges within the Tennessee
Valley, moreover, the TVA agriculturists have found it necessary to
take the side of the system of land-grant colleges as a whole, in national
terms. This necessity, coupled with the personal history of the partici-
pants and the working relationship with the national leadership of the
Farm Bureau, 20 has made them acutely aware of the over-all controversies
which have been outlined in the preceding pages. Consequently, in ex-
ercising delegated discretionary powers, and in taking advantage of
opportunities within the Authority for influencing policy on tangential
matters, the agriculturists have sought to commit the Authority as a
whole to the cause espoused by the Farm Bureau and the extension
service. This has been done not in speeches or testimony or other forms
of public witness, but through the "dry" method by which the weight
of an organization is wielded : the establishment of precedents, the ac-
ceptance or rejection of opportunities for joint effort, the subtle intima-
tions of disaffection put on paper, and the more plain verbal injunctions
which permit field agents to know the attitudes and desires of head-
quarters.
The handling of proposals for cooperative relations among agencies is
one of the touchstones by which real attitudes may be determined, and
is a field within which discretion normally has wide play. "Cooperation"
is a term which remains unanalyzed so long as its use in administrative
parlance neglects to specify references, procedure, and consequences. A
long list of agencies with which cooperative relationships are maintained
may be as meaningless as it is casually impressive so long as it represents
an amalgam of the simple with the complex and slurs over the real and
significant distinctions among the relationships established. This sub-
stitution of shadow for substance is a convenient means of fulfilling the
formal demands of a statute or a Congressional committee, as well as
the need for a shield against criticism of some specific relationship.
23 The Agricultural Adjustment Administration has played an important role in
the distribution of TVA fertilizers, but apart from the purchase of fertilizer en bloc
there has been little in the way of a close organizational relationship. It appears
likely that the decided advantage offered by AAA in providing a convenient outlet
for surplus fertilizer materials outweighed any considerations of interagency con-
troversy.
29 See above, pp. 141-145.
166 University of California Publications in Culture and Society
Insofar as cooperation bespeaks more than words though these too
are important it has meaning for the future of the organizations in-
volved. Especially when negotiation is infused with motives which go
beyond the substantive programs in question, that future is at stake. If
an official is hostile to the organization with which cooperation is pro-
posed, he will consider the possibility that the joint program will serve
to strengthen that organization through : (1) the precedents to be estab-
lished, which may be difficult to overcome at a later date ; (2) the possible
influence of that organization upon his own rank and file ; (3) the extent
to which funds allocated to him may leave his control in the course of
the joint activity; (4) the danger that public credit may be shared, or
that the two organizations will become confused in the public mind;
(5) the possibility that access to client publics, not previously available,
will now be afforded to the cooperating organizations; and (6) the
establishment of an actual machinery (personnel, facilities, reciprocal
commitments) which might become intrenched. 80 If the attitude is
friendly rather than hostile, and especially if there is interest in actually
building the cooperating organization, then precisely these considera-
tions will strengthen the case for cooperation. But if the official is
organizationally self-conscious, he will not ignore them.
It is not the least of the paradoxes characterizing the TVA that there
should have been so little in the way of rapprochement with the Farm
Security Administration. Both agencies seemed to be part of the more
experimental phase of the New Deal, and both derived support from
the liberal-labor movement in the United States. It was perhaps not
unreasonable to expect, therefore, that the two agencies would have a
very close relationship, each contributing to the other's strength and
prestige. But this is to reckon without the split in the character of TVA :
carrying out its mission in the field of electric power, but permitting
the agricultural program to be controlled by representatives of the ex-
isting institutions of the area. In fact, as we have seen, the actual alli-
ance of the TVA agriculturists was with the bitterest enemy of FSA
the Farm Bureau. Consequently, the Authority as a whole was placed
in the anomalous position of rebuffing what many would have considered
to be one of its natural allies.
The policy of rebuffing proposals for cooperation with FSA stemmed
from the Agricultural Relations Department of TVA. Other divisions
of the Authority were willing, even eager, to work with the FSA. But
30 These considerations are general, and apply as well, of course, to the relations
between political parties, for example, and other nongovernmental groups of all
kinds.
Selznick : T VA and the Grass Roots 167
constant pressure from the agriculturists, whose control over the major
subject-matter department interested in rural people placed them in a
strategic position, led to a virtually complete elimination of FSA from
joint activity with the Authority. This objective was accomplished in
rather subtle ways, so that no complete defense or explanation has been
spread upon the record. It was made easier since no formal relationships
on a general scale had ever been initiated, for the extension service
group had taken control over the agricultural program from its incep-
tion. Consequently, no open break was necessary, nor any extended
discussion within the Authority. Nevertheless, it is possible to set forth
some details on the methods used by the agriculturists to swing the
TVA away from cooperation with the Farm Security Administration.
1. The attitudes of leading personnel within the Agricultural Rela-
tions Department are obliquely hostile to the Farm Security Adminis-
tration and other federal agencies in agriculture which do not follow
the grass-roots method. This is expressed in general terms, as in talks
to personnel stressing the dangers of increased centralization and the
threat to local institutions from new ventures by the federal govern-
ment ; outspoken support of the Farm Bureau position during days of
national controversy that the county agent should be placed in charge
of all agricultural programs at the county level ; and more rarely, rela-
tively explicit references such as that of the Director of the Department
that he was "scared of 'security' spelled with a large V " Considering
also the attitudes outlined above, 31 it is hardly surprising that the TVA
agriculturists would be cold to proposals for cooperation with FSA.
2. Officials of the TVA Forestry Relations Department were eager
to work out a joint program with the Farm Security Administration,
partly because of the resistance felt by foresters on the part of the exten-
sion service. Some contacts were initiated in 1938. By that time, however,
the TVA foresters had already been bracketed with the Agricultural
Relations Department under the supervision of the Chief Conservation
Engineer. In November, 1938, a discussion was held including repre-
sentatives of the foresters, the Director of Agricultural Relations, and
the Chief Conservation Engineer, at which the question was posed of
whether TVA "should recognize Farm Security." According to one
source, Director of Agricultural Relations McAmis said "No," to which
the Chief Conservation Engineer added, "Whatever McAmis said on
the question must go." More recently, an attempt was made to route
assistance to FSA through the State Forester in Tennessee, and some
31 See above, p. 112.
168 University of California Publications in Culture and Society
work was done on FSA farms in this way, but a TVA forestry official
was told to "lay off" because of the extension service attitude : "If you
want to love me, you can't love FSA."
3. In 1941, an official in TVA central management initiated a proposal
that the Authority undertake a general understanding with Farm Se-
curity Administration whereby joint effort might be concentrated on
the problems of migratory farm labor, surplus farm population, and
the reaching of progressively lower economic strata of farm families.
It was suggested by this official that the approach of TVA and FSA
were the same. This proposal was rejected by the TVA agriculturists,
in a vein which called into question the validity of the FSA enterprise
as a whole and insisted that the methods of the two agencies in agricul-
tural matters were quite dissimilar, thus precluding any general under-
standing. As in other instances of this kind, the agriculturists pointed
out that FSA clients were free, as individuals, to participate in the TVA
extension service program. This formula, which shunts the problem of
interagency cooperation from the level of organizational relationships
to that of participation by individual clients, avoids the organizational
implications of cooperation and in effect does not constitute meaningful
cooperation. The planned programing of a joint enterprise is a far cry
from a permitted overlap of clienteles. 32
4. While the FSA was carefully insulated from the major phases of
the TVA agricultural program, a point of official contact between the
two agencies was established in 1937. This involved participation of the
FSA in the work of aiding some of the families displaced by the TVA
reservoirs. The Relocation Section of the Reservoir Property Manage-
ment Department did not share the anti-FSA bias of the agriculturists
and was willing to make recommendations to central management favor-
able to official cooperation with FSA. In connection with this work, a
representative of FSA was provided with office space at TVA headquar-
ters to function as an informational coordinator. The agriculturists ob-
jected to this arrangement from the beginning, demanding that any
arrangement with FSA should include the land-grant colleges as a
party. 33 However, the FSA representative remained. But upon the occa-
32 This is the same sort of problem as a "united front" in politics. If a joint political
meeting is proposed by one organization to another, it may be suggested by one if
it really wishes to go ahead on its own that "there is no need for organizational
alignments, anyone may come to the meeting." Such a reply will not be accepted in
good part by officials who must pay attention to the problem of party recognition,
joint leadership, etc.
33 The tactic of introducing the land-grant colleges as a necessary party, with which
the TVA agriculturists countered any suggestion that the Authority enter into
Selznick : T VA and the Grass Roots 169
sion of a routine resignation of the designee, the agriculturists moved to
thwart the appointment of a successor. This was accomplished in a way
which revealed the high order of self -consciousness which the agricul-
turists brought to bear upon all matters of organizational relationship.
The Chief Conservation Engineer, acting as spokesman for the agricul-
turists,
. . . advised in considerable detail of the efforts that were now and for the past few
months had been under way in the Department of Agriculture at Washington to
develop a procedure under which the programs and activities of the action agencies
like Farm Security could be coordinated with and channeled through the extension
agencies in each state. He advised that an arrangement was being or had been devel-
oped under which a committee would be established in each state, with the State
Extension Director serving as chairman of the committee. The Bureau of Agricul-
tural Economics was to serve as the coordinating agency and would furnish the
secretary to each state committee. [The Chief Conservation Engineer] said he would
like to see the Authority cooperate in the development and maintenance of the over-all
plan rather than to ask a special arrangement be provided through which to carry
on the cooperative relationships. 3 *
In this way, the agriculturists introduced matters of broad national
policy into the determination of specific problems. It would be impos-
sible to understand this move without the knowledge, developed above,
of the commitment of the TVA agriculturists to the extension service
wing of the Department of Agriculture. The viewpoint of the Chief Con-
servation Engineer was accepted, and the TVA request for an FSA
representative was withdrawn. It seems doubtful that FSA appreciated
the Authority's concern lest it "cause the FSA to deviate from the policy
and procedure recently developed in order to make its facilities available
to displaced families."
THE SOIL CONSERVATION SERVICE CRISIS
The basic pattern of administrative cooperation envisioned by the TVA
agricultural leadership found its expression in a three-way memoran-
dum of understanding uniting the Department of Agriculture, the
TVA, and the land-grant colleges of the seven Valley states. 35 This
agreement established a Correlating Committee, and included the fun-
damental provisions that all proposals for coordinated activity on the
cooperative arrangements with FSA, was often repeated, and served as an effective
way of hinting that the FSA was "not wanted," or at least, if there had been (as
there was not) a serious challenge, as a delaying maneuver.
84 At a meeting, March 3, 1939, which included the General Manager, the Chief
Conservation Engineer, and representatives of the TVA Reservoir Property Manage-
ment Department.
33 See above, pp. 95-98.
170 University of California Publications in Culture and Society
part of any of the parties would be channeled through the joint com-
mittee. Viewed in its broader context, the agreement and the machinery
it established functioned to unite the old-line elements in agricultural
administration against the new agencies which, it was feared, might
overwhelm and displace the established institutions. At the time of the
execution of the memorandum of understanding, in 1934, the repre-
sentatives of the several agencies involved shared the same basic outlook,
so that unity was real rather than merely verbal. However, as the admin-
istrative offspring of the New Deal in agriculture gained strength and
stability, there was observed a weakening of the support given by the
Department of Agriculture to the three-way understanding in the
Tennessee Valley area.
The push and pull within the USDA was reflected in the changing
attitudes of the Department toward earlier administrative arrangements
and, more concretely, in the ascent of new personnel, with divergent
outlooks, to high posts within the Department. Thus, as the Department
changed, its role as party to the memorandum of understanding neces-
sarily changed. The earlier informal meaning of the agreement, which
presumed a unified outlook on the part of the participants, became a
source of embarrassment to newly powerful elements within the Depart-
ment. This change was reflected in the membership of the Correlating
Committee when Milton S. Eisenhower, Land-Use Coordinator, became
the representative of the Department. Eisenhower tended to represent
the "new" Department and was not wholly sympathetic with the out-
look and aims of the TVA agriculturists. As a result, the real content
of the TVA-USDA-co] leges pattern of cooperation was seriously altered,
despite the continued existence of the written memorandum and the
formal procedures of cooperation. The deterioration of the old relation-
ship, from the standpoint of the TVA officials, was highlighted in 1941
by a critical situation which almost resulted in an open break between
the Department and the Authority, and which caused the latter to
make explicit the long-run organizational consequences which a change
in the established procedures might invoke. This crisis situation (or
"casual breakdown") 88 casts into bold relief the implications of the
grass-roots policy for the place of the Authority within the federal
administrative system.
^ The crisis situation in the individual presents "a short and dramatic dislo-
cation of his usual relationships with any given social institution or social pattern,"
and "reveals problems which exist in ordinary life and which, in ordinary life, do not
carry sufficient emotional reaction to lift them into the field of awareness." See J. S.
Plant, Personality and the Cultural Pattern (New York : The Commonwealth Fund,
1937), pp. 58-59.
Selznick : T VA and the Grass Roots 171
Among the new agencies which departed significantly from the ad-
ministrative method of the agricultural extension service is the Soil
Conservation Service. This organization was established in 1933 within
the Department of the Interior as the Soil Erosion Service, under the
authority of the soil erosion control provisions of the National Indus-
trial Recovery Act. Under the leadership of a dynamic and evangelistic
former member of the Bureau of Chemistry and Soils, Hugh H. Bennett,
the agency grew to one of the largest of the action organizations within
the fields of agriculture and soil conservation. The SES early under-
took a radical departure from the methods approved by other farm
agencies, especially in the practice of supplying labor and materials
to farmers without cost. This direct approach to farmers with "some-
thing to offer beside advice and inspiration" placed the extension serv-
ices in a disadvantageous position and was, moreover, inimical to the
basic philosophy of the older organizations, which stressed the need for
substantial contribution by the farmers themselves to any conservation
program. In addition, the establishment of a new channel of approach
to the farmers was a threat to the dominance of the American Farm
Bureau Federation, which lobbied in Washington to divert the func-
tions of the SES through the extension service machinery, However,
the leadership of the SES was convinced that it had a special mission
to perform, requiring a measure of zeal and dedication which could
not be preserved if it did not have control over the execution of its soil
conservation program. This attitude of dedication to soil conservation
was transmitted through the ranks, serving to weld the organization
into a coherent group and infusing it with a distinctive idealistic spirit.
As time went on, some of the earlier practices of the SES were modified,
taking into account that other agencies existed with which stable rela-
tions had to be maintained, but its existence as an independent organi-
zation, with direct access to the farmers and soil conservation problems,
was not compromised.
The Soil Erosion Service was transferred to the Department of Agri-
culture in March, 1935, and, after approval of the Soil Conservation
Act of that year, the SES was designated the Soil Conservation Service.
In its new form but still under Bennett's leadership the organization
entered a period of expansion. With the revamping of the legal foun-
dations of the Agricultural Adjustment Administration in 1936, soil
conservation was firmly established as a basic aspect of the action re-
sponsibilities of the federal government, and the SCS was given a key
role in the administration of the Soil Conservation and Domestic Allot-
172 University of California Publications in Culture and Society
merit Act. The original objectives of the SES, centering on the preven-
tion and control of erosion, were considerably broadened in the new
responsibilities of the SCS. In 1940, the basic objectives of the Soil
Conservation Service were broadly formulated :
... to aid in bringing about desirable physical adjustments in land use with a view
to bettering the general welfare, conserving natural resources, establishing a per-
manent, balanced agriculture and reducing the hazards of floods and siltation.
In attaining these objectives the Service, as an action agency, provides technical
and material assistance in determining and making physical adjustments for the
conservation and proper use of soil and water resources and in connection with
farm forestry and water facilities development. The Service also engages in sub-
marginal land purchase and development. 37
As the functions of the SCS were extended, and as the intensity of its
activities increased, problems of relationship with other agencies having
similar functions grew in importance. This situation assumed a peculiar
form in the Tennessee Valley, shaped by the special role of the TVA and
its Agricultural Relations Department.
During the early days of the Soil Erosion Service and the TVA, some
preliminary moves toward joint activity particularly in relation to
a possible soil-erosion experiment station in the Valley were under-
taken, chiefly by representatives of the A. E. Morgan wing of the
Authority. Nothing came of these efforts, however, and in 1935 an
official statement of the Department of Agriculture defined the relation
of the Soil Conservation Service to the TVA. A report submitted to
the Secretary of Agriculture on June 5, 1935, by his Committee on
Soil Conservation recommended:
That the Soil Conservation Service not undertake erosion-control work in the area
under the jurisdiction of the Tennessee Valley Authority except as agreed upon
by the existing TVA coordinating committee of three representing the Department,
the Tennessee Valley Authority, and the seven States in the Tennessee River Basin.
This TVA committee has been established to provide coordination of all agricultural
activities conducted within that area. 38
This procedure, which amounted in practice to the virtual exclusion of
the SCS from soil conservation work in the Valley, was accepted as
basic policy until 1940. Up to that time, the SCS went through a period
of great expansion, but refrained from entering those counties of Ten-
37 H. H. Bennett, Chief, Soil Conservation Service, and M. L. Wilson, Director of
Extension Work, "A policy Statement on Kelationships between the Soil Conserva-
tion Service and the Extension Service," March 29, 1940 (mini.).
38 U. S. Department of Agriculture, "Eeport of the Secretary's Committee on Soil
Conservation," June 5, 1935 (mini.). This report was approved by Secretary Wallace
on the following day.
Selznick : T VA and the Grass Boots 173
nessee, Georgia, North Carolina, Alabama, Kentucky, Mississippi, and
Virginia which were denned as within the basin of the Tennessee River.
One of the chief means of this expansion was the enactment of soil
conservation laws by the state legislatures establishing local soil con-
servation districts. These districts were authorized to request aid from
the Soil Conservation Service, which had a statutory obligation to re-
spond. The state laws were often known as "SCS Bills" and it was gen-
erally reported that SCS officials were involved in lobbying activities
at the local level, to insure participation by the state in the program.
In 1940, after the Kentucky legislature passed such a bill, the SCS is
reported to have entered the Tennessee Valley portion of the state, but
a really critical problem did not develop until the following year, when
it was precipitated by the developing situation in Alabama.
Early in June, 1941, the Northeast and Northwest Alabama Soil
Conservation Districts applied to the Soil Conservation Service for
inclusion in its regular program of assistance to such districts. This
was the first formal issue to be raised envisioning a change in the exclu-
sion order of 1935 which restrained the SCS from working within the
Valley area. The matter was placed before the Correlating Committee, 39
resulting in a two-year period of negotiation and discussion which made
explicit the underlying organizational assumptions which the TVA had
made a part of its grass-roots approach in agriculture.
The significance of the SCS invasion of the Tennessee Valley area
was immediately apparent to the TVA agricultural leadership. In July,
1941, within a month of the application of the Alabama districts for
SCS aid, the TVA Director of Agricultural Relations had laid before
his colleagues a detailed exposition of the organizational consequences
of the new development. These consequences were seen in the broad
context of changing policy and principle within the agricultural organ-
izations of the United States government. Noting the development of
"increasingly wide divergences of point of view" between the Depart-
ment of Agriculture on the one hand and the Authority and the land-
grant colleges on the other, this official called for a general clarification
of relationships among the interested agencies, including such farm
organizations as the American Farm Bureau Federation. It was sug-
gested that the TVA Board of Directors press for the recognition by
the Department of Agriculture of the idea of the Tennessee Valley as
"an experimental area for unified development under decentralized
39 Dean Thomas P. Cooper, University of Kentucky Extension Director ; Director
J. C. McAmis, Agricultural Relations Department, TVA ; Land-Use Coordinator Mil-
ton S. Eisenhower, U. S. Department of Agriculture.
174 University of California Publications in Culture and Society
administration"; that is, under the primary guidance of the TVA,
working with and through the state and local agencies. The issue was
phrased in strong terms: the Authority must either capitulate to or
aggressively oppose new forces and principles which necessarily would
lead to the defeat of the Authority's multipurpose function. Moreover,
the actual disturbance of existing relationships among agricultural
agencies operating in the Tennessee Valley was seen to have reached
such a critical stage as virtually to force upon the TVA the necessity
to come to some over-all decision whether it would acquiesce in the new
changes of policy or refuse to do so. In either case, far-reaching con-
sequences would have to be predicted and faced.
The TVA agricultural program (so the argument ran) is rooted in
a basic decision that the resources of the land-grant college system
within the Valley be utilized to the full, and no independent TVA
field force be established. At the time this decision was made, this re-
flected the dominant administrative philosophy of the U. S. Department
of Agriculture. But gradually the TVA agriculturists noted a change
in the attitude of the Department. It was observed that the Department's
original support of the program based on the memorandum of 1934
weakened as its newer action agencies increased in strength and influ-
ence. At the same time, disagreements arose between the Department
and the land-grant colleges with respect to the allocation of program-
matic functions and responsibilities. Concomitantly with changes in
viewpoint, a rapid turnover of administrators within the Department
relegated the three-way Valley agreement to the limbo of the forgotten
and the unenf orced.
The new proposal of the Department, that the SCS enter into agree-
ments with the Northeast and Northwest Alabama Soil Conservation
Districts, was not acceptable to the TVA officials. They maintained
that such a move would involve the Authority in relationships not
envisioned by the memorandum of understanding. In effect, this could
only mean that the TVA so interpreted its obligations as to force it
to insist that the extension service had an exclusive right to execute
an agricultural program within the Valley area. No attempt had been
made by the TVA to mobilize the resources of the new federal agencies
in agriculture or to fit them into a regional agricultural program. Nor
could the Authority do so and yet insist upon the exclusive organiza-
tional legitimacy of the extension services. Moreover, the TVA was in
this instance objectively carrying out the program of the American
Farm Bureau Federation which, as we have already noted, was like-
Selznick: TVA and the Grass Roots 175
wise opposed to the Soil Conservation Service. Nor did the TVA con-
ceive of the SCS situation as a minor or passing affair : It was considered
evident that the new approach of the Department of Agriculture, as
reflected in its support of the SCS proposal to work in northern Ala-
bama, constituted a threat to the TVA's basic concept of a decentralized
regional agency. Consequently, the issue at hand could not be considered
in isolation, but rather in terms of its general meaning for the course
of the entire regional program.
The TVA agriculturists flatly believed that, when the Alabama issue
was brought before the Correlating Committee, the Department of
Agriculture's proposal should be rejected. But the consequences which
such a course of action involved were not, from the viewpoint of TVA,
altogether desirable. If the Authority were to reject the Department's
proposal, the latter might simply withdraw from the memorandum of
understanding. Then the Department might proceed to muster all its
resources for an all-out program within the Valley area, possibly in-
ducing personnel on the TVA-extension test-demonstration program to
leave for better-paying positions ; it might even stimulate and sustain
whatever local opposition there might be to the Authority's electric
power policy. Pressure to forego earlier loyalties to the TVA might be
brought to bear upon state institutions under threat that flexibility in
the use of federal funds would be curtailed. With this line of reasoning,
the TVA agriculturists implied that the TVA might lose an effective
base of support for its whole program, especially in electric power,
unless the problem of relationships in agriculture were solved in such
a way as to preserve the original pattern. More generally, it was sug-
gested that the Department of Agriculture might use the Authority's
rejection of its proposal in Alabama to support the contention that the
establishment of regional agencies such as TVA is administratively
undesirable, especially in relation to the possible congressional approval
for the inclusion of the Cumberland River basin within the jurisdiction
of the TVA. 40 Or, further, the Department might withdraw from its
agreement to utilize TVA phosphates in the AAA program, resulting
in serious embarrassment to the Authority. Finally, if, in the face of
vigorous action by the Department in the Valley, the TVA should con-
tinue to press its own program, the result would be widespread confu-
sion and duplication which would not long be tolerated by Congress.
In the event of congressional action against the TVA on this issue, the
40 It will be recalled (p. 78) that this issue did in fact arise during 1941, and that
the Department of Agriculture took its position on organizational grounds.
176 University of California Publications in Culture and Society
dismemberment of the Authority, long desired by its enemies, would
be on the road to accomplishment.
But the Authority's dilemma in this crisis was completed in that,
should decision be made to acquiesce in the proposals of the Depart-
ment, the consequences might not vary greatly from those of a decision
to oppose them. The alternative was phrased in these terms :
If the Department should intensify its programs within the Tennessee Valley under
its own direct methods, which the cooperating state institutions would be forced to
accept just as they have been forced to do in the areas outside the Tennessee Valley,
the Authority would be forced, with the allegiance of the institutions to TVA thus
alienated, either to withdraw from the agricultural field or to initiate its own pro-
gram directly. To withdraw means the dismemberment of TVA, the very thing
which the enemies of TVA have advocated for many years. To proceed directly
with its own program would violate the principles for which the Authority has
long stood. It would increase enormously the expenditures in the area through the
prosecution of duplicating programs. This the Congress would not countenance, and
properly so.
In fact, it seems to have been realized within the Authority that the
agency could not hope to escape completely from the pressure of these
harsh alternatives, and that the situation could be saved if at all only
by avoiding a head-on clash with the Department of Agriculture. Some
reconciliation was necessary which would preserve the form of the old
relationships and limit as far as possible the actual extension of SCS
operations in the Valley. However, while such an administrative clash
would have to be avoided, it was considered possible that the Author-
ity might muster its political resources in order to effect a favorable
resolution of the situation. This orientation was expressed thus :
It is imperative, therefore, that the Authority and the Department come to terms
and that the Department return to its allegiance to the memorandum of under-
standing. The first step in reconciliation of ideas and objectives would seem to be
for the Board of Directors of the Authority, either alone or with representatives of
the land-grant colleges, to approach the Secretary of Agriculture to determine:
1) His understanding of the relationships of the Department to the Authority;
and 2) His position with reference to the existing memorandum of understanding.
In the event he is found to be unfavorable to the principles of the memorandum,
this fact should be known to the Authority and to the cooperating institutions, who
would then inform agricultural leaders and farm organizations throughout- the
country and particularly in the Tennessee Valley, so they then might inform their
state representatives in Congress and request that the matter be resolved by the
President of the TJ. S. or in the Halls of Congress.
It is not known whether such pressure was actually brought to bear at
the highest political levels, but it is known that a direct effort was made
Selznick : T VA and the Grass Roots 177
to enlist the leadership of the American Farm Bureau Federation on
the side of the Authority in this situation. Moreover, a special meeting
of the presidents of the land-grant colleges of the Valley states was
held in June, 1942, with the participation of Director H. A. Morgan
of the TVA Board. This meeting was called expressly to show the
solidarity of the land-grant institutions with the TVA and its approach.
The TVA found itself in a difficult position with relation to the col-
leges. Its own commitment to the grass-roots method had made the
continued cooperation of the colleges indispensable, but the colleges
could not restrict their consideration to the effect of new patterns upon
the TVA as an organization. This was exhibited in the dilemma of the
Director of the Alabama extension service who was caught between his
sympathy for the TVA and his inability to refuse the concrete help
which SCS was in a position to give to Alabama farmers. He could not
accede to pressure from the TVA to oppose the entry of the SCS into
the Valley portion of the state, particularly as that portion represented
only a small part of the state's total area, a consideration which must
have been important for all of the seven Valley states except Tennessee.
However, in the course of the meeting of college presidents in June,
1942, the TVA was able to gain (or perhaps salvage) the following
commitments from the institutions:
1. That, although they recognized the administrative obligation of the Authority
to uphold the regional program, they asserted the concern of the institutions in their
own right in a program of unified development as a means of fulfilling their insti-
tutional obligations. . . .
2. That the responsibilities of the Department, the institutions, and the TVA
with respect to the regional program were so clear and binding that, regardless
of the merits of the North Alabama issue, none of the parties should withdraw
from the program. Failure to cooperate would endanger the success of the national
experiment which the Congress and the Administration had set up in the public
interest.
The TVA was wholly dependent upon the continued cooperation of
the state institutions in its agricultural program, so it was imperative
to mobilize their support for the continuation of the memorandum of
understanding. It was well understood by the TVA officials that the
Authority could not hope to work out its relationships with the land-
grant colleges independently of the Department of Agriculture. Hence
the continued existence of a three-way agreement was considered
indispensable.
At a meeting of the Correlating Committee on September 4, 1941,
the general situation was reviewed, but far from any agreement being
178 University of California Publications in Culture and Society
reached, there ensued a difference of opinion over even the conclusions
reached at the meeting itself. In any case, immediately upon his return
to Washington, Land-Use Coordinator Milton S. Eisenhower extended
his approval, on behalf of the Secretary of Agriculture, to the general
agreements executed between the Northeast and Northwest Alabama
Soil Conservation Districts and the USDA's Soil Conservation Service.
Eisenhower took the position that conditions had changed since the
exclusion order of 1935 and necessitated a revision of the earlier pro-
cedures. The Department officially recalled that the 1935 functions of
SCS consisted largely of research, employment of relief labor, and the
management of watershed demonstration projects in a relatively small
number of areas. The demonstration approach of the SCS had now
changed to one of extensive assistance to the states, with the objective
of reaching the greatest possible number of farmers. Congress had
appropriated funds for this work, on the assumption that no admin-
istrative barriers would be placed in the way of assistance to all
farmers eligible under the law and requesting it through the local soil
conservation districts. Under pressure of these conditions, it was felt
that the Department could no longer feel justified in excluding the
SCS from operation within the Tennessee Valley.
At the same time, the Department was willing to compromise in
relation to the TYA program to the extent of enjoining the SCS to take
into account the special circumstances within the Tennessee Valley:
(1) The SCS was to "refrain from carrying on promotional work aimed
specifically at the organization of soil conservation districts within the
Valley"; (2) the SCS was to avoid working on any farm in the TVA-
extension service test-demonstration program "without first obtaining
the concurrence of the local officer in charge of such TVA demonstra-
tions" ; 41 (3) services to the soil conservation districts were to be carried
on in close cooperation with extension service personnel working in
the districts.
In the course of the negotiations, the Department stressed that in
fact it had not generally operated through the Correlating Committee
within the Valley area. With the exception of SCS, the action programs
of the Department had been carried on directly within the Tennessee
Valley just as in other parts of the country. These programs included
those of the Agricultural Adjustment Administration, the Farm Secur-
ity Administration, the Farm Credit Administration, and the Rural
41 It had earlier been charged that SCS activities in Kentucky had involved a con-
scious eff ort on the part of SCS officials to "sign up" TVA demonstration farms.
Selznick : T VA and the Grass Roots 179
Electrification Administration. This was often ignored by the TVA
officials, who persisted in the assertion that the SCS crisis revolved
about a generally new approach by the USDA in the Valley. In fact,
the memorandum of understanding did not have the controlling force
which the TVA agriculturists wished to give it, though it was used as
a weapon in support of TVA's constituency commitments whenever
possible.
The Department did not recede from its position that the exclusion
order of 1935 with respect to SCS would have to be abandoned. In
1942, after a protest by the chairman of the Correlating Committee,
the Secretary of Agriculture served notice that the memorandum of
understanding would be terminated. However probably after further
negotiations at the top executive level among the agencies involved a
new memorandum of understanding was signed in September, 1942,
which included the following statement as the only significant change
in the earlier agreement :
It is understood that neither this memorandum of understanding nor the Correlating
Committee provided for herein, has administrative jurisdiction which would prevent
any signatory agency or institution from carrying out the provisions of legislation
for which said signatory is responsible. 42
This new language permitted the Department to carry forward its SCS
activities without challenge by the Correlating Committee, since it was
held that upon request assistance to the soil conservation districts was
required by law. At the same time, the over-all framework of the
agreement now even formally reduced to little in the way of regional
integration beyond the relation of TVA and the colleges was main-
tained as a gesture to the continuity of the old relationships.
In this chapter we have examined one phase of the consequences of
the exercise of discretionary power at the grass roots. The commitment
of the TVA agriculturists to the land-grant college system involved the
agency as a whole in discordant relationships with the federal admin-
istration. Intervention in the Valley occurred during a period of gen-
eral controversy in the field of agricultural administration, heightening
the self -consciousness of all participants. Hence this analysis provides
a detailed illustration of the basis for the Authority's failure to adjust
itself to the going federal administrative structure.
42 "Statement and Memorandum of Understanding between U. S. Department of
Agriculture, Tennessee Valley Authority, and Agricultural Colleges of Alabama,
Georgia, Kentucky, Mississippi, North Carolina, Tennessee and Virginia." Effective
October 16, 1942. Signed by the Presidents of the Institutions, the Chairman of TVA,
and the Secretary of Agriculture (mini.).
CHAPTER VI
f
UNANTICIPATED CONSEQUENCES, 2:
LAND-USE POLICY AND THE
CHARACTEB OF TVA
The purchase of a minimum amount of land around TVA reservoirs would provide
opportunity for private enterprise to exercise its initiative in the development of
the waterfront free from the stifling effect of public ownership and in accord with
the public demand. THE TVA AGRICULTURISTS (1941)
THERE is A VAGUE and ill-defined quality which, unacknowledged and
often poorly understood, represents a fundamental prize in organiza-
tional controversy. This is the evolving character of the organization
as a whole. What are we ? What shall we become ? With whom shall we
be identified? Where are our roots? These questions, and others like
them, are the special responsibility of statesmen, of those who look
beyond the immediate context of current issues to their larger implica-
tions for the future role and meaning of the group. To pose these ques-
tions is to seek more than the technical articulation of resources, methods,
and objectives as these are defined in a formal program or statute. To
reflect upon such long-run implications is to seek the indirect conse-
quences of day-to-day behavior for those fundamental ideals and com-
mitments which serve as the foundation for loyalty and effort.
"Consequences," writes Dewey, "include effects upon character, upon
confirming and weakening habits, as well as tangibly obvious results."
And in an earlier passage from the same work: "Character is the
interpenetration of habits. If each habit existed in an insulated com-
partment and operated without affecting or being affected by others,
character would not exist. That is, conduct would lack unity being only
a juxtaposition of disconnected reactions to separated situations." 1
These considerations from the social psychology of the individual pro-
vide us with tools for organizational analysis. Organizations, like indi-
viduals, strive for a unified pattern of response. This integration will
define in advance the general attitudes of personnel to specific prob-
lems as they arise. This means that there will be pressure within the
organization, from below as well as from above, for unity in outlook.
As unity is approximated, the character of the organization becomes
defined. In this way, the conditions under which individuals may "live
1 John Dewey, Human Nature and Conduct (New York: Modern Library, 1930),
pp. 46, 38.
[181]
182 University of California Publications in Culture and Society
together" in the organization are established, and a selective process
is generated which forces out those who cannot identify themselves
with the evolving generalized character of the organization. The evolv-
ing character, or generalized system of responses, will be derived in
large measure from the consequences of day-to-day decision and be-
havior for general patterns of integration.
An examination of the logic of this development provides us with
a theoretical link between the concept of the character of an organiza-
tion and that of administrative discretion as discussed above. 2 The act
of discretion permits the administrator to introduce considerations
tangential to the formal or stated objectives of the organization. At
the same time, by the accretion and integration of modes of response,
the officialdom is able to invest the organization with a special character.
This special character tends in turn to be crystallized through the pres-
ervation of custom and precedent. It is further reinforced by the selec-
tive process which rejects those members who cannot fit in, and shapes
the personal orientation of those who remain or who are recruited.
In a situation charged with conflict, the process of discretion will be
subjected to close scrutiny, and the quality of administrative decision
will tend to be infused with a high degree of self -consciousness. The
scrutiny of the opposition and the self -consciousness of the leadership
will alike center upon the question of commitment. What attitudes and
what symbols are commanding the loyalties of the staff? What prece-
dents are being established? What alliances are being made? Such
issues will be uppermost in the minds of leading individuals during
periods when the evolution of the character of an organization is not
yet settled. The possibility of stating that some given line of action is
the "settled policy of this organization" is one of the strategic objec-
tives in such a controversy. Or, in a field somewhat oblique to questions
of policy, there may be conflict over the "heroes" of the group. Lauda-
tory references to a set of individuals as the "fathers" of the organiza-
tion's policy and outlook may help to define the accepted antecedents
of the group ; as a result, a whole series of doctrinal commitments are
inferred from those antecedents, though not necessarily formally in-
cluded in the program or objectives of the organization. In such cases,
controversy may occur over the question of whether some individual's
memory is to be celebrated in the official newspaper or bulletins, or
whether certain slogans and symbols, traditionally associated with one
general tendency or another, will be included. Conflicts over apparently
2 See above, pp. 64 ff .
SelznicJc : T VA and the Grass Roots 183
minor matters of this sort are typically aspects of the struggle to deter-
mine the character of an organization.
The internal organizational pressures which drive toward a unified
outlook and systematized behavior receive their content, or substantive
reference, from the play of interests and the flow of ideas which charac-
terize the organization's social environment. In this way, the internal
process of character formation though generated by needs which
may be ref errable primarily to the organization as such comes to be
stamped with the typical hallmarks of its own historical period. The
general commitments and attitudes of the organization (i.e., its char-
acter) will tend to crystallize around value problems which are current
in the environment. For an organization whose discretionary power on
social questions is broad, there is pressure to make a choice among the
"historical alternatives" that are available. Once that choice is made,
the organization will tend to reflect in its own character the general
sentiments with which it has become aligned.
The struggle over the outcome of this process may extend over a
long period, and may be compromised from time to time, but the stake
is always all important. The attempt to define an organization's char-
acter cannot be divorced from the struggle for leadership or from the
possibility of internal convulsion. It is precisely in the struggle over
an evolving organizational character that a given leadership having
certain personal qualities most easily becomes the receptacle of a social
ideal. As such, that leadership incumbent or proposed is conceived
as indispensable to the goal of stamping the organization as a whole
with the desired ideal. A leadership can become indispensable when it
has convinced itself and its constituents that some alternative elite
cannot be trusted with the exercise of character-defining discretionary
powers ; whereas the possibility of adequate replacement in the execu-
tion of formal executive functions is not normally in doubt.
These preliminary remarks permit us to round out the logic of our
thesis, which may be stated in these summary terms :
1. The theory of grass-roots administration as reflected in the idea
of "working with and through local institutions" remains unanalyzed
prior to an explication of its concrete application and consequences.
2. In the major field of the TVA agricultural program, the exercise
of discretion in applying the grass-roots concept initiated a constituency
relationship involving the land-grant college system on the one hand
and the Agricultural Relations Department of TVA on the other.
3. The constituency relation may be viewed as a mechanism of the
184 University of California Publications in Culture and Society
informal cooptation of grass-roots elements into the policy-determining
structure of the TVA.
4. The impingement of the administrative constituency upon policy,
operating through its organized representatives within the Authority,
takes advantage of the special prerogatives of the latter group. These
prerogatives derive from its status as an integral part of the formal
structure of the Authority.
5. The prerogatives of the Agricultural Kelations Department in-
clude the exercise of discretion within its formally assigned subject-
matter field as well as the exertion of pressure upon the evolution of
general policy within the TVA as a whole.
6. Especially in a situation marked by controversy, the selective con-
sequences of the pressure exercised by the TVA agriculturists signifi-
cantly affects the role of the TVA as a whole within the federal system.
7. As major character-defining issues come to the fore, this pressure
influences the evolving character of the Authority as a whole.
8. Consequently, our search for the meaning and significance of the
grass-roots policy verifies the hypothesis that the channeling of admin-
istrative responsibilities through previously existing institutions result
in consequences for the role and character of the initiator which were
not anticipated by the formal theory or doctrine. We also learn some-
thing about the mechanics of this process.
These points are demonstrable on the basis of the empirical analysis
presented in this study. However, it should be mentioned that there
are specific matters of fact, for which adequate direct evidence is not
available, which are considered significant by informed participants.
These include (1) the special relation of the Agricultural Relations
Department of TVA to Dr. H. A. Morgan, one of the three members
of the TVA Board, with the implication that the TVA agriculturists
had in effect a direct representative on the Board of Directors a priv-
ilege enjoyed by no other single department in the Authority; and
(2) the political importance of the extension service machinery, spread
throughout the Valley, as enhancing the power of the agriculturists
within the Authority again the only department within TVA having
the potential power of wielding an organized apparatus in the interests
of policy questions which it might consider crucial. 8 These considera-
3 It would seem that the Personnel Department, in its relation to the A. F. of L.
Tennessee Valley Trades and Labor Council, might have a similar opportunity, how-
ever. Doubtless much depends on the relative weight of the various constituencies
in the area of operation and on the relative loyalties of the several departments to
the organization as a whole.
Selznick: T VA and the Grass Roots 185
tions, among others, help to explain the relatively dominant position
of the agricultural group, resulting in a degree of influence upon policy
which is not seen from an inspection of the formal structure of the
Authority.
The thesis thus stated concentrates attention upon organizational
dynamics. It is necessary to bear in mind, however, that the evolution
of policy may be traced to broader and more general factors than the
specific pressures exerted by groups within and around a given organ-
ization. In a sense, the forces generated by the process of organization
per se may be thought of as the means by which the pressing but
more general imperatives of the particular historical period are given
effect. Such motives as prestige and survival may adequately im-
pel action; but in general the need to rally forces broader than the
small group for whom these motives are effective will make necessary
an appeal based on moral or political principles which can be defended
on their own level. In this way, the organizational struggle is provided
with doctrinal content and a socially acceptable arena. But the price
of that strategy involves a commitment to a set of ideas or interests.
Hence those ideas and interests are provided with a means of interven-
tion, a driving force which they may not be able by themselves to
generate, but which, once generated, can carry them along with it.
To this qualification we may add another : changing historical con-
ditions may seriously affect the choices available to discretionary power.
In an administrative agency, controversies over policy may become
academic overnight if statutory changes occur. Or the range of choice
may be gradually restricted by changing economic conditions as from
a period of depression to one of prosperity, or of relative plenty to rela-
tive scarcity as well as by shifts in the political climate which may
make certain choices less expedient at one time than another. As these
changes occur, the relative strength of forces within the organization
may vary, reflecting needed reorientations. As the new realignment
takes place, policy will change, but it might be rash not to look beyond
the inner-organizational controversies for the cause of the change. In
general, it is suggested that these considerations which link specific
organizational pressures to the more general imperatives and forces
of the time will temper any tendency which may arise from a concen-
tration upon organizational dynamics to ignore the goals and demands
provided by a particular historical period.*
4 Daniel Bell, in conversation, has made the cognate point that power and the
ends of power may not be divorced in a proper sociology. This is an important re-
minder for all who study the mechanics of organizational interaction.
186 University of California Publications in Culture and Society
TVA AND PUBLIC LANDS
We may now turn to a case history in the evolution of administrative
policy which carries forward our search for the consequences of the
grass-roots doctrine, and at the same time exemplifies the general prin-
ciples which have been set forth above. This policy is within the prov-
ince of discretion of the Board of Directors of the TVA. It defines the
extent to which the Authority engages in a program of land acquisition
for public purposes incidental to a major project: the construction of
a series of dams and the consequent impounding of large bodies of water
along the course of the Tennessee River and its tributaries.
The term "incidental" used here is of special significance. Both the
theoretical and the practical interest in this record stem from the fact
that this is an area of discretion available to the Board. In 1941, the
general manager of TVA stated that the problem of determining the
boundary ("taking line") around the reservoirs, within which land
was to be purchased, represented a major area of difficult administra-
tive coordination within the agency. But this kind of difficulty can arise
within a field of decision only when the latter is not rigidly determined
or exhausted by technological considerations. The attainment of statu-
tory objectives, when a measure of leeway or choice within even a lim-
ited sphere is allowed, opens the door to differences as to incidental or
tangential goals. These may be assigned a formal collateral status and
thus be given due consideration in the execution of general policy. As
we have suggested, controversy over these collateral goals may in effect
place the character of the organization at stake.
As with other "general-welfare" or developmental objectives, the legal
power of the TVA to acquire public land for general purposes is se-
verely limited ; here, too, the context of the Authority's origin and the
broad conceptions of its proponents express a fuller set of objectives
than does the Act itself. President Roosevelt's message to the Congress
of April 10, 1933, had a prophetic ring :
It is clear that the Muscle Shoals development is but a small part of the potential
public usefulness of the entire Tennessee River. Such use, if envisioned in its en-
tirety, transcends mere power development : it enters the wide fields of flood control,
soil erosion, afforestation, elimination from agricultural use of marginal lands, and
distribution and diversification of industry. In short, this power development of
war days leads logically to national planning for a complete river watershed involv-
ing many states and the future lives and welfare of millions. It touches and gives
life to all forms of human concerns. 5
5 House Doc. 15, 73d Cong., 1st sess. (1933). Emphasis supplied.
Selznick: TVA and the Grass Roots 187
The preamble to the Act was likewise broadly phrased :
To improve the navigability and to provide for the flood control of the Tennessee
River; to provide for reforestation and the proper use of marginal lands in the
Tennessee Valley; to provide for the agricultural and industrial development of
said valley ; to provide for the national defense. . . . 6
Despite, these broad affirmations of purpose with respect to a public
land policy, the provisions of the Act which followed the enacting clause
neither established machinery nor delegated authority which would
give effect to those ambitions. The Authority was assigned power to
acquire real estate "for the construction of dams, reservoirs, transmis-
sion lines, power houses, and other structures, and navigation projects
at any point along the Tennessee River, or any of its tributaries," and
a condemnation procedure was established. In addition, Sections 22 and
23 of the Act authorized studies, experiments, and demonstrations, as
well as recommendations to Congress which would aid in the proper
use, conservation, and development of the natural resources of the
Tennessee Eiver drainage basin and adjoining areas. However, these
general-welfare provisions of the Act are regarded as among its weak-
est parts, and the management of the agency has steadily curtailed
efforts to base substantial parts of the program upon them.
Owing to these weak statutory delegations, the Authority has fol-
lowed the policy of executing what it considered to be its general-
welfare responsibilities in an oblique way : essentially by investing the
administration of those programs which did have a sound legal foun-
dation with certain ideals of public policy. Thus the TVA accepted
responsibility for the increased malarial hazard which resulted from
the creation of new lakes and backwater areas; the creation of con-
struction villages was conceived as generating responsibility for the
provision of social services, including library facilities; the effect of
newly impounded reservoirs upon local communities justified efforts
to aid in the establishment of town planning services ; and the need to
protect the public investment in dams and reservoirs was linked to a
general interest in soil conservation on the basis of the theory that
siltage could be effectively forestalled by minimizing soil erosion in
the drainage basin. These considerations, and many others, effectively
distinguish the Authority's administrative methods from those of rela-
tively irresponsible private corporations. At the same time, however,
the discretionary element in administration is vastly enlarged. To the
extent that such tangential responsibilities are accepted, without clear-
6 48 Stat. 58 (May 18, 1933). Emphasis supplied.
188 University of California Publications in Culture and Society
cut definitions of policy in the statute, the possibilities of variant
interpretations are present. "The ideals of public policy" may then
depend upon the specific interests and backgrounds of those actually
charged with the day-to-day administration.
In an agency such as TVA, land policy in acquisition and in man-
agement is a criterion to discern the generalized attitudes of .respon-
sible administrators on significant issues of social policy. The purpose
for which TVA has acquired tracts of land in the course of its general
operations have been listed by a TVA official as follows :
Essential or unavoidable acquisition: to acquire reservoir areas necessary for
impoundment or flooding; to acquire those areas necessary for the construction of
dams, power structures, and for the development of related improvements and op-
erations; to control reservoir margin uses not consistent with the purposes of the
TVA Act; to consummate transfer of the properties of certain power companies
whose holdings and operations have been taken over by the Authority.
Desirable but not essential acquisition: to avoid uneconomic severance of property
lines through over-purchase of entire tracts; to eliminate excessive costs of re-
construction and relocation of roads, bridges, and other improvements; to provide
opportunities for development of public projects such as recreational areas, experi-
mental grounds, game refuges, fish hatcheries and ponds, tree nurseries, etc.; to
provide for the protection of the reservoirs from excessive silting and surface run-off
through public control of adjacent lands.
Those purposes listed as "desirable but not essential" and to some extent
those listed as "essential or unavoidable" are subject to the discretion of
the TVA Board. It is evident, moreover, that the issues involved are of
such a nature that long-debated judgments of value must be brought
to bear in setting policy within the agency. The role of government in
such matters as the provision of recreational facilities and the conser-
vation of natural resources, as more generally the obligation of govern-
ment to capture for public use the increments of value brought about
by technological changes, represents one of the character-defining motifs
of the present historical period. This is especially true in those areas of
the United States which have been least influenced by such ideologies
of the past two generations as the conservationist movement and the
ideal of the positive state. Considering the often-remarked persistence
of pioneer attitudes toward the land in the Southern states, it is not
surprising that the issue of land policy should have been subject to
prolonged debate within the TVA organization. It is important for this
analysis to discern the mechanics of that debate, and the pressures in-
volved, as these are related to the organizational consequences of the
official grass-roots doctrine.
Selznick: TVA and the Grass Roots 189
Before tracing the development of the land-purchase policy within
TVA, some reference should be made to the viewpoint on this question
adopted by public agencies with which the Authority is normally iden-
tified. The following statement, from the first issue of the Bureau of
Agricultural Economies' Land Policy Review, is indicative :
The first major change of direction in land policy was initiated by the conservation
movement which attained momentum in the early years of the twentieth century. . . .
The conservation movement, by instituting public ownership as a means of protect-
ing national economic resources, marked a radical departure from the laissez-faire
economic policy of the nineteenth century. . . . The second great contribution to the
stream of land policy . . . has been the development of methods whereby a sound use
of land could be achieved on private as well as public lands. [This has marked a
change] in our traditional concepts of land tenure. . . . We are forced to recognize
the existence of a social interest in the private ownership of land and to protect that
interest where a recalcitrant individual owner fails to do so. 7
This point of view, which emphasizes the emerging recognition of a
public stake in land utilization, is supplemented by the conception of
public land acquisition as an instrument of achieving a sound national
land-use program. A recent formulation of a general program of land
acquisition was undertaken in 1940 by the National Resources Planning
Board, and motivated in these terms :
In the light of present-day conditions, then, private ownership of most of our land
resources is generally recognized as a basic tenet. However, public land acquisition
of some lands now in private hands is justified, not only for public buildings, wildlife
refuges, recreational sites, and for purposes of conservation and restoration of forest
resources, but also as a means of assisting in preventing or correcting maladjustments
in land use that stand in the way of the fullest and most effective utilization of our
national resources. 8
The NRPB pointed out that programs of public aid to agriculture have
apparently relied too much on the potentialities of improved farm man-
agement. In its view, however, "there are . . . many lands naturally so
unproductive or in such unproductive operating units, that even the
extensive public use of adjustment aids now available, plus financial
grants, merely enable occupiers to hold on a little longer. Consequently,
land acquisition has to-day become one instrument, and a basic one,
among many instruments for the effective conservation, development,
and wise use of the Nation's resources." 9 The principles embodied in this
7 L. C. Gray, "Our Land Policy Today," Land Policy Eeview, Vol. 1, No. 1 (May-
June, 1938).
8 U. S. National Eesources Planning Board, Public Land Acquisition in a National
Land-Use Program, Part I: Rural Lands (Washington: U. S. Government Printing
Office, 1940), p. 2.
Ibid., p. 3.
190 University of California Publications in Culture and Society
general policy are widely believed to be those which motivate the TVA,
in common with other agencies presumed to stand in the conservationist
tradition. However, the Authority's leadership now explicitly states that
there is a widespread misconception of the TVA's character in this re-
spect. In fact, the evolution of the Authority's policy on land acquisition
and land use is a history of a slow but definite change in outlook, involv-
ing a significant and conscious reversal of major policy.
CONSERVATION GOALS IN THE EARLY PERIOD
During the first years of the TVA organization, the "superidealists" as
they came to be known in later days held many strategic positions
within the agency, and attempted to shape its character seemingly con-
sistent with the spirit in which the new enterprise was launched. This
group included those most closely associated with Arthur E. Morgan,
whose program for social change within the Valley brought him into
sharp conflict with Harcourt A. Morgan and the latter 's ally on the
Board, David E. Lilienthal, within a few months of the founding of the
agency. The sharp split within the Board was soon reflected at the lower
administrative echelons in hostility and controversy over many issues,
among which the question of land policy loomed large both in importance
and in persistence.
The first major construction project undertaken by the TVA was the
erection of Norris Dam on the Clinch River, one of the large tributaries
feeding the Tennessee. The policy which determined the amount of land
acquired in the course of this operation marks a distinct phase in the
history of the Authority. Acquisitions in the Norris area included an
"overpurchase" of approximately 120,000 acres land lying above the
contour which would be reached by the newly created reservoir. Of
course these lands, which included vast tracts of forest cover, lay off the
main stream of the Tennessee River. It could hardly be expected that
acquisition on such a scale would be possible at dam sites along the Ten-
nessee itself. But significant is not so much the extent of the overpurchase
at Norris, as the viewpoint within the Authority during that period.
The most aggressive proponent of extensive acquisition was Edward
C. M. Richards, Chief Forester of the TVA from 1933 until 1938. Rich-
ards was a close associate of Arthur E. Morgan, then chairman of the
Authority, and is usually included on the roster of superidealists whose
departure is said to have left the agency on firm, realistic ground. He
was bitterly opposed from the beginning by the agricultural group. But
such factors as the support of the chairman who was also Chief Engi-
Selznick: TV A and the Grass Roots 191
neer and primarily responsible for construction of the dam in addition
to that of the old Land Planning and Housing Division; the general
spirit of aggressive idealism which prevailed during the early days ; the
economic depression which stimulated an interest in the retirement of
submarginal lands ; and the general character of the Norris area which
seemed to be especially suitable for a large-scale conservation enter-
prise a ii combined to give the position of the Chief Forester a domi-
nance which it would not long hold. Richards attempted to commit the
Authority to an interest in the public management of forest lands on an
extensive basis, carrying over the experience with such projects which
he had gained in Europe. But his concern for the land reached beyond
forestry. Thus, in 1934, in the course of a discussion of the evolving land-
purchase policy of the Authority, he stated :
It has been my conviction for a long time that there is a very great need for the
TVA not only to discuss some basic policies but to come to some definite conclusions
and commit itself to definite policies. Particularly is this true, I believe, in connection
with the matter of forest, critical erosion, and marginal land areas within the Valley.
All areas falling within these three classifications are of great importance to the
Tennessee Valley project because in the first place they are today being used in such
a way as to hold down the standard of living of the rural population of the Valley and
to increase the soil erosion ; and on the other hand, they are not being used in such a
way as to constructively build up their potentialities of service in the future.
These three types of land forest, critical erosion, and marginal represent a con-
siderable proportion of the total area of the Tennessee Valley. Only in the case of
forest lands is any serious attention being given to the care and future development
of such areas. And what is being done in the case of forest lands is entirely inadequate,
not only at the present time but as far as plans for the future are concerned
In the case of critical erosion areas and marginal lands, the need for discussion and
definite commitment of policy by the Board is in one way even more imperative and
pressing than in the case of forest lands, for the reason that for the most part more
people are endeavoring to live and earn a livelihood on such areas than on forest lands.
The results of such effort are an impoverished rural population and a constantly
deteriorating physical condition of the lands themselves.
This point of view, insofar as its goals were concerned, was essentially
that of such other governmental agencies as the Soil Erosion Service, the
Bureau of Agricultural Economics, and the National Resources Com-
mittee.
Richards likewise struck out at the tendency within the Authority to
think of soil erosion control as deriving its value solely from the func-
tion of preventing sil ting-in of the dams, rather than as a vital objective
for its own sake. 10 In view of later changes in outlook, its is important
10 We have noted above, pp. 58-59, that such conceptions as "water control on the
land" serve to tie general welfare objectives (e.g., soil conservation) to the consti-
192 University of California Publications in Culture and Society
to make clear this early approach within the TVA to the role of a public
agency in land-use policy. Continuing the discussion just quoted, Rich-
ards stated :
I believe that soil erosion control is a part of land use. Proper land use is of vital
importance of the TVA to help the population of the Valley, to stabilize local indus-
try, and to increase the use of electric power. Proper land use calls for the cessation
of destructive practices on the one hand and the initiation of constructive produc-
tive use on the other. Such proper land use is often impossible financially in the case
of the owners of small tracts of land in hilly sections. They must make their own
living off of the land they own even though this may tend toward destroying their
own property through erosion.
To make possible the proper use of marginal lands, reforestation and erosion con-
trol by some public agency is in many cases essential. In order to do this, however,
ownership of the land in fee is necessary. This is for the reason that complete control
of what is done on the lands necessitates full ownership. 11
The last two sentences are of special significance, for it is precisely the
principle of full ownership in fee simple which the Authority has pro-
gressively discarded.
In general, the goals set forth by the first Chief Forester were sup-
ported by those within the Authority most closely associated with its
planning objectives. The head of the Land Planning and Housing Divi-
sion, Earle S. Draper, was generally associated with Richards during
the early period, though his own professional interest in afforestation
as such was more limited. Draper supported the proposals for exten-
sive land acquisition in the Nor r is area because of "the definite sub-
marginality of the land, the extremely low level of income, the lack of
opportunity, and the danger of increased settlement to perpetuate low
living standards." The professional planners were accustomed to incor-
porating criteria of social utility as a part of practical judgment, and
had long been committed to public ownership and control as a basic
instrumentality for avoiding the misuse of marginal areas. The agricul-
turists within the Authority were deeply committed to an opposing view-
point, and hotly contested the criteria proposed by the planners and
tutional pegs of navigation and flood control. The need for oblique justifications of
developmental programs has been strong because of the relatively weak delegations
in the statute. This tendency to dissemble was, apparently, one of A. E. Morgan's
objections to the administrative methods of the majority bloc. Moreover, in respect
to Eichards' strictures, it should be recalled that during the early days of the Author-
ity some controversy arose between those who conceived of the project primarily as
an electric power business and those who placed its developmental functions in the
forefront. To the extent that developmental functions were chargeable to power
operations, this would increase the cost of production and thus endanger the "yard-
stick" objectives of electric power operations.
u See also E. C. M. Eichards, "The Future of TVA Forestry," Journal of Forestry,
XXI (July, 1938), 643-652.
Selznick : T VA and the Grass Roots 193
the foresters. But it was not until the resolution of the internal crisis,
resulting in the elimination of A. E. Morgan and many of his followers,
that the agriculturists were able to exert a disproportionate influence
outside their own special province.
THE RESERVOIR PROTECTIVE STRIP
The broad objectives of Chief Forester Richards, quoted above, char-
acterize the first phase of TVA policy on land acquisition. A second
phase, for more limited acquisition, is marked by the explicit acceptance
by the Board of the policy of establishing a buffer belt of publicly owned
land around each reservoir. This policy, analogous to the objective of
creating green-belt areas around urban communities, was consistent
with the professional outlook of the planning group and found support
through a convergence of the interests of several subject-matter depart-
ments within the Authority.
Until 1942, the TVA's policy on land acquisition was formulated sepa-
rately for each reservoir in a resolution adopted by the Board. Thus,
on December 20, 1935, the Board approved a resolution on land purchase
for the Guntersville Reservoir (on the main stream of the Tennessee
River) which read in part as follows :
WHEEEAS, the Authority desires to start at once the purchases of land in the
Guntersville Beservoir area, and
WHEEEAS, the Board considers it necessary to purchase a protective belt of land
around the reservoir,
BE IT EESOLVED, That the Director of Engineering Service is hereby authorized
to approve for purchase the lands to be flooded by the Guntersville Beservoir, and in
addition, such lands as are necessary to form a protective belt around the Gunters-
ville Eeservoir, extending in general from three hundred (300) feet to one thousand
(1,000) feet from the edge of the reservoir, with occasional variations in width for
short stretches as are dictated by topography, ownership, or other valid considerations.
The explicit acceptance of the reservoir protective strip as official policy
continued, with some variations, until 1941. A belt of publicly owned
land 300-1,000 feet in width was also authorized for Chickamauga Reser-
voir, near Chattanooga, in 1936. A modification was introduced in 1939
with respect to Kentucky Dam, at which time a "reasonable" protective
strip was authorized without specifying dimensions. In addition, other
criteria for justifiable overpurchase were more specifically delimited:
to eliminate peninsulas or tips projecting into the reservoir ; to eliminate
the danger of erosion and consequent silting by the purchase of steep
hillside slopes adjoining the reservoir ; to eliminate severance damages
by the purchase of inaccessible or worthless farm remnants, islands, and
194 University of California Publications in Culture and Society
peninsulas ; and to avoid the necessity of relocating roads or other facili-
ties when it would be more economical simply to purchase relatively
inaccessible areas that would have to be provided with access and other
services.
The resolution on Watts Bar Eeservoir behind another main stream
dam adopted February 28, 1940, eliminated the earlier references to
the Board's general interest in the purchase of a protective belt around
the reservoir (as in the "whereas" clause quoted above), but did author-
ize the provision of "a reasonable protective strip around the margin
of the reservoir." Impractical severance of individual properties and
uneconomical provision of access facilities were again cited as justifica-
tions for overpurchase, but a new element was introduced in a provision
that the right to flood (flowage easements) might be acquired in lieu of
fee simple titles "on such lands as are occasionally flooded, where such
easement purchases appear practicable and economical." This sequence,
while still accepting throughout the principle of a reservoir protective
strip, indicated a gradual weakening of Board policy. The weaker policy
was transformed into an outright reversal when the gathering strength
of those who opposed public ownership in principle became decisive in
shaping the policy of the TVA Board.
During this second phase of the land acquisition policy, the several
interested departments of the Authority took positions on the issue
which reflected their various professional commitments. Those in favor
of the policy of providing a protective belt around the reservoirs in-
cluded :
The professional planners. After the departure of B. C. M. Richards,
and the consequent truncating of the authority of the foresters, the
Regional Planning Studies Department took the lead in favoring public
ownership of a protective strip as a matter of principle. This group bore
much of the brunt of the internal debate, and formulated the principle
issues in terms which reflected the broad views of many planning officials.
One such formulation was set forth by the leading TVA planner of the
first seven years, Earle S. Draper, in 1938 :
In the sense that the building of our reservoirs may have broad relationships and
serve as a means to the end of social and economic improvements, the acquisition of
reservoir purchase area may relate to regional adjustments as well as to actual water
control in such a way as to make rather important demonstrations of the best uses
of land. The actual ownership of the protective strip may give indirect benefits in
demonstrations relating to farming, forestry, recreation, etc., of great importance
to the over-all objectives of the TVA program.
Selznick: TVA and the Grass Roots 195
The planning group underwent a series of changes which resulted in a
general curtailment of objectives and a "more realistic" outlook, but
continued their efforts in support of the protective strip policy until the
actual reversal by the Board was consummated. 12
The foresters. Although adopting a more limited perspective on
land acquisition and land-use policy, the foresters continued to support
the idea of public ownership as that was applied in the protective strip
policy. The Forestry Relations Department, even under its new leader-
ship, tended to be aligned consistently with the planners in defending
the protective strip policy. Moreover, in 1940, Chief Forester Willis M.
Baker (who succeeded E. C. M. Richards) stressed the need for the
Authority "to recognize more clearly certain fundamental principles
and basic objectives relating to land use/' Though generally recognized
as more moderate in his views than his predecessor, he too was faced with
the problem of overcoming pioneer attitudes toward the land. In the
course of a discussion of this problem, Baker said :
In the formulation of a policy based upon local and landowner responsibility, several
facts must be kept clearly in mind. The past depletion of lands and natural resources
may be attributed largely to our national conception of the rights of the private
citizen and to national policies set up to protect these rights even at the expense of
public welfare. Our American assumption has always been that private initiative,
through self-interest, would find ways of keeping land productive. We now discover
that this same self-interest, together with lack of concern for the public or the future,
has caused the ruin of many lands and the exhaustion of certain resources, resulting
in serious social and economic problems. Conditions have changed greatly since the
original land policies were adopted. The public is now faced with a situation that
demands realization and acceptance of responsibility for remedial action.
The foresters found it necessary to carry on a continuous effort to coun-
teract those forces within the Authority notably the agriculturists
who maintained the continued cogency of older assumptions concerning
public land policy.
The Maps and Surveys Division. The protective strip policy found
additional support from this organization, under the jurisdiction of the
Chief Engineer, which had the responsibility of making on-the-spot
determinations of the amount of land to be purchased within limits set
by the Board of Directors. This group supported the protective strip
13 These changes included some overturn in personnel, a change in title of the
Eegional Planning Studies Department by omitting the word "Planning," an aban-
donment of the concept of over-all planning for the Tennessee Valley analogous to
that carried on for the nation by the National Resources Planning Board, and the
loss of authority to conduct planning studies -in rural land use. The course of the
regional planning group inside TVA is not unlike that which led to the curtailment
of the functions of the NEPB and its eventual elimination.
196 University of California Publications in Culture and Society
policy for technical reasons, rather than as a matter of principle. Thus
surveying economies might be effected if deviations from the irregular
shorelines of the reservoirs were permitted ; or it might often be con-
sidered advisable to buy more land than that to be covered by water in
order to avoid impractical severance, i.e., the purchase of only part of a
farm, leaving the most unproductive upland part in private ownership.
Since problems of this sort would normally arise, it was considered
desirable and economical to permit a reasonable amount of overpurchase.
Possibly the social outlook of the chief of this division, who had been
associated with A. B. Morgan in the early years of the agency, was a
factor beyond purely technical considerations, which influenced the
position of the Division on the protective strip policy. The group was
placed in a strategic position, since for a long period the Board delegated
to the Maps and Surveys Division the authority to approve the purchase
of specific tracts of land. It is significant that in 1942, concomitantly
with a general abandonment of the protective strip policy, the Division
was deprived of its authority to exercise discretion in this respect.
The Reservoir Property Management Department. Although direct
evidence of its importance is lacking, some participants suggest that this
department, charged with administering TVA-owned lands, also threw
support in favor of the protective strip policy, "in order to protect their
reason for being." The abandonment of the protective strip policy not
only reduced the extent of TVA-owned lands which would require man-
agement, but also carried the threat that the Authority would sell its
holdings and thus eliminate much of the department's functions.
The groups so far mentioned constituted the major alignment in favor
of the protective strip policy. Other departments, less directly concerned,
tended to support the existing policy. But the opposition was powerful,
and ultimately prevailed. It included :
The agriculturists. As TVA operations moved from the tributaries
to the main stream of the Tennessee River, and concomitantly with the
overturn of personnel which followed the removal of A. E. Morgan in
1938, the TVA agriculturists opened a campaign to effect a reversal
of the agency's land-purchase policy. This program included, as will be
noted below, a fundamental opposition to the principle of public owner-
ship of land. In pursuing this policy, the agriculturists were reflecting
the attitudes and interests of the agricultural institutions of the Valley
area.
The Chief Engineer. After the removal of A. E. Morgan, who had
exercised a dominant influence over the construction program, the agri-
Selznick : T VA and the Grass Roots 197
culturists found a welcome ally in the person of a new Chief Engineer,
T. B. Parker. Apparently lacking some of the broader interests which
characterized the construction leaders of the early period, Parker intro-
duced an economy-minded approach, with general opposition to policies
which would add to the cost of the construction program, including the
overpurchase of land around the margins of the reservoirs. The organi-
zational weight of the Chief Engineer in exercising discretion and in
influencing the development of Board policy, was thrown to the side
of the agricultural group. The latter understood the convergences of
interests involved and readily admitted that, in effect, an intraorgan-
izational alliance had been consummated. The ascendancy of the new
economy-oriented leadership among the construction engineers is a sig-
nificant symptom of the modification of the character of TVA. It meant
that those charged with administering an important phase of the pro-
gram were no longer dedicated to the use of the technological changes
introduced by the agency as instruments for the achievement of tangen-
tial social purposes. In a minor way, the Maps and Survey Division
long under the influence of A. E. Morgan's point of view underwent
an experience similar to that of the Forestry Relations Department.
Since the Maps and Surveys Division was within the formal jurisdiction
of the Chief Engineer, pressure could be put upon it to make an adjust-
ment to new emphases in policy.
The divergences of opinion and interest reflected in the departmental
alignment outlined above found expression within a short time fol-
lowing the organizational crisis, congressional investigation, and ad-
ministrative reorganization of 1937-1938. A debate on land-use and
land-purchase policy was initiated by 1939 and continued until 1942,
at which time the Board established a new policy, reversing earlier posi-
tions and climaxing a steady trend away from the conservationist objec-
tives and methods associated with such agencies as the Forest Service,
Soil Conservation Service, Bureau of Agricultural Economics, and Na-
tional Resources Committee. The nature of the debate, and its sequence,
is shown in a series of general statements formulated by the several
interested groups within the Authority :
1. In June, 1939, after a year's experience in the Authority, Chief
Forester Baker expressed his serious concern with regard to the TVA's
policy on problems of acquisition, use, and management of rural lands.
He voiced in a more moderate and restricted vein some of the disagree-
ments which his predecessor had made public. In addition to stating his
own views, Baker noted that the Authority was receiving criticism in
198 University of California Publications in Culture and Society
Washington which expressed "the unfavorable attitude of many repre-
sentatives of conservation agencies and national associations toward
what they regard as the anticonservation activities and practices of the
Authority." He suggested that, on the basis of his own experience, there
appeared to be some justification for this criticism, requiring the formu-
lation of definite policies and more efficient procedures by the Authority.
While proceeding cautiously and apparently with an eye to the strength
of the agriculturists within the agency, Baker supported the need for
land acquisition in the interests of such purposes as recreation, game
refuges, fish hatcheries, tree nurseries, and other public projects. He
also obliquely raised the question whether the Authority was not ignor-
ing or overlooking some of its responsibilities in its zeal for dependence
upon private initiative and for the protection of local interests. 13
2. In November, 1939, the Director of the Regional Planning Studies
Department (later redesignated simply Regional Studies Department)
expressed the belief that the Authority's organization for land-use plan-
ning was inadequate. In particular, he observed that the Department of
Agricultural Relations did not seem able to execute its responsibilities
for agricultural land-use planning. He hinted broadly that the agricul-
turists did not cooperate with other departments in joint studies of prob-
lems involving agricultural planning, and suggested that the official
formulation of the duties of the Chief Conservation Engineer over-
stressed the agricultural aspect and did not give enough weight to for-
estry. These matters have already been touched upon above. Significant
in this context is that this summary assignment of responsibility for
purported defects represented part of the debate over land-use policy,
stressing its administrative dimension.
3. In February, 1940, the Chief Conservation Engineer formulated
his views on the Authority's land-acquisition policy. Accurately reflect-
ing the views of the TVA agriculturists, he based his proposals for a
general land policy on the following propositions :
1. Land can be acquired only "to carry out the purposes of the Act." This should
be practically accomplished with maximum economy to the Authority and with the
least disruptive effect to the region, taking into consideration fairness to the land-
owner whose individual property is affected.
2. Where there are elements of doubt as to the amount of land necessary for statu-
tory purposes, these doubts should be resolved in favor of leaving the land in private
ownership. This is particularly important to the agricultural industry upon which the
impact of land removal for reservoir use is necessarily greatest.
13 See the report below of the land-use controversy involving the Biological Survey,
pp. 206 ff.
Selznick : T VA and the Grass Roots 199
The proposed policy definitely cast aside the collateral objective, favored
by the planners and others within the Authority, of capturing for the
public the increment of value resulting from the construction of reser-
voirs. In general, the agriculturists have considered the grass-roots
policy to be the prime collateral objective of the Authority's activities.
In the land-acquisition policy, the grass-roots method was conceived of
as demanding minimal purchase at the reservoir margins. Moreover, it
was proposed that lands acquired by the Authority for reasons of ulti-
mate economy (farm remnants and inaccessible and isolated areas)
should be disposed of by sale as rapidly as possible instead of trying
"to find TVA uses for these properties as reasons for continuing to own
and operate them." At another occasion during the same period, a repre-
sentative of the Department of Agricultural Relations put the matter
in categorical terms : his department wanted private ownership of land
to extend to the water's edge.
4. After considerable discussion, the whole problem of the protective
strip was thoroughly reviewed when determining land acquisition policy
for the Fort Loudon project in 1941. In May of that year the opposing
groups within the Authority made most explicit the varying principles
which grounded their judgments and consequently the basis for the
enduring disagreements. It was clear that the issue would not be resolved
except by some formal step taken by the Board of Directors. Conse-
quently, the General Manager requested the staff to formulate the diver-
gences of opinion for review by the Board.
Provided with this opportunity the chief proponents of the reservoir
protective strip and their opponents formulated the principle bases of
their respective positions. The protective strip policy was identified as
follows :
a) Land lying below the contour "at the top of the gates," i.e., subject to flooding
due to the backing up of water behind the dam, but not the maximum backwater line,
would be acquired by outright purchase in fee simple, except where there was clear
warrant for the purchase of easements.
b) A protective strip extending back approximately 300 feet from this contour
would be provided.
c) Other lands lying between the contour at the top of the gates and the backwater
line are acquired by easements except where purchase in fee simple appeared clearly
desirable.
d) Fee purchase would be authorized when necessary to avoid impracticable sever-
ances, isolation, and road relocation. 14
14 These criteria had been earlier accepted for purchase at Watts Bar Dam, with
the exception of the width of the protective strip which in that case over the objec-
tion of the planners had been set at 50 feet. In formulating their general policy,
200 University of California Publications in Culture and Society
The arguments marshaled by the planning group in support of this
policy reflected the viewpoint of those who were willing to use govern-
ment intervention as a tool for the achievement of public welfare objec-
tives incidental to the primary purposes of the agency and subject to its
discretion. Four major arguments were set forth by the proponents of
the protective strip policy :
i) The protective strip would capture for the public the incremental
value created by the new reservoirs an increment which would other-
wise accrue to only a limited number of individuals. This was conceived
to be consistent with the policy of recognizing the public interest in an-
other by-product of reservoir construction electric power. The cost
of capturing this increment would be sufficiently low to make it simply
incidental to the primary program of the Authority; failure to accept
this responsibility would leave the agency open to the justifiable charge
of neglecting its rightful obligations to the public. Moreover, the cost of
recapturing reservoir sites once a project was completed would normally
be prohibitive. Hence a future and foreseeable obstacle to the creation of
parks and other projects would be avoided by the reservation of a pro-
tective strip. 15
ii) The protective strip, creating a reservoir area of publicly owned
waterfront land, would permit the realization of opportunities for public
benefit as they might arise. Though public demand may be slow to
crystallize, the technical judgment of a responsible public agency such as
TVA should be able to take advantage of foresight in the interests of the
general welfare. The agriculturists within the Authority had demanded
that land be acquired only for specific purposes for which a current need
was evident. The principle of a reserve area to be utilized as occasion
arose was formulated in order to counter that demand.
iii) The protective strip would permit the Authority to establish con-
ditions at the reservoir margins which would provide opportunities for
private enterprise to operate the shoreline facilities soundly and profit-
ably but without detriment to the public interest. The Authority would
at once protect the continued cleanliness and beauty of the reservoir
those in favor of the protective strip were asking for one more generous than that
accepted for Watts Bar.
15 Various examples of increases in land values on the chain of lakes created by
the reservoirs were cited, including the following: at Guntersville, Alabama, an
increase for some tracts of as much as 5,000 per cent since the creation of Gunters-
ville Lake; near Caryville, Tennessee, the donation of 150 acres of land for park
purposes was thwarted when a large portion of the area was withdrawn for a resi-
dential subdivision; at another park site, the cost of supplementing TVA purchases
by two small additional tracts of land was materially greater than the cost of similar
tracts within the TVA strip acquired before the filling of the reservoir.
Selznick : T VA and the Grass Roots 201
sites and at the same time provide opportunities, through licensing and
leasing of public lands, for private investment. It was pointed out that
at Guntersville, Alabama, under TVA ownership of the commercial
and industrial reservoir frontage, three oil terminals, a sawmill, a sand
and gravel company, a grain elevator, a lumber yard, and an automobile
transportation terminal had been established within a three-year period.
iv) The protective strip would permit public control of lands at the
reservoir margins, with the opportunity of enforcing land-use objec-
tives : the prevention of abuses of land and forest cover and the exclusion
of undesirable developments having inadequate standards of safety,
sanitation, operation, and appearance. It was stressed that public owner-
ship of the reservoir strip could compensate for the absence of legal
control at the local level, and remain the most reliable device for avoid-
ing shacks, billboards, and other structures which would deteriorate
the scenic resources of the waterfront lands. Lack of a protective strip
in some areas was said to have made possible the development of sub-
standard services by uncontrolled private operators and the overde-
velopment of facilities which, if subsequently abandoned, may create
backwater slums.
It is apparent that the reasoning of the proponents of the protective
strip policy in general accords with the philosophy which underlay the
establishment of the Authority itself. On the other hand, the opponents
of the protective strip appear to have been indifferent to such considera-
tions. The policy they supported during the 1941 discussion was :
a) Land below the backwater line, but above the minimum operating level of the
reservoir, would not be purchased outright. Flowage rights would be acquired through
the purchase of easements.
b) Lands required for authorized programs and projects of the Authority would
be purchased in fee.
c) Lands purchased to avoid excessive costs of highway relocation, isolation, or
impracticable severance would be sold concurrently.
The protagonists of this policy based their program upon a set of as-
sumptions which cannot be considered consistent with the broad objec-
tives expressed by President Koosevelt and others who initiated the TVA
project. Moreover, it tended strongly to pull the TVA outside of the
conservationist movement as that was expressed in the activities and
organizations established since the pioneer efforts of such men as Gifford
Pinchot. 18 Opponents of the protective strip policy stated their objec-
tions.
10 In 1943, the general manager explicitly recognized that TVA was not within
the traditional conservation movement. As formulated by the agriculturists this
202 University of California Publications in Culture and Society
i) The purchase of a minimum of land "would leave in private control
the maximum land suitable for use in the established agricultural econ-
omy in an area where it is more urgently needed than in any other region
of the nation." In effect, this argument rejected the conception that
the introduction of technological changes, such as the impounding of
reservoirs, inevitably altering established conditions, would offer oppor-
tunities for the achievement of corollary social objectives. No account
was taken of accompanying changes that, would be introduced by pri-
vate owners, such as the establishment of residential subdivisions along
the new shorelines.
ii) Minimum land purchase at the reservoir margins would reduce
displacement of population and expenditures for readjustment activi-
ties. This was one of a series of economy-oriented arguments, which in
effect rejected the earlier assumption of responsibility by the Authority
for the social consequences of technological change. The burden of caring
for population remnants would be shifted to local agencies in many
cases inadequately equipped to handle them. While a short-term advan-
tage might be gained, the long-run consequences of avoiding the problem
of displacement would ultimately have to be faced in some way.
iii) For this analysis, perhaps the most significant of the arguments
proposed in an official way by the agriculturists was that which affirmed
that the policy of minimum land acquisition "would provide the oppor-
tunity for private enterprise to exercise its initiative in the development
of the waterfront free from the stifling effects of public ownership and
control and in accord with the public demand." At one stroke, the oppo-
sition of the agriculturists to public ownership of land was laid bare.
This meant that a change in policy based upon such an assumption would
be not merely a matter of expediency but would cut at the fundamental
social outlook, hence at the character, of the TVA as a whole."
A REVERSAL OF POLICY
The steady retreat from the policy of providing a protective strip in
public ownership around the reservoirs was reflected in an increasing
pressure to purchase flowage easements rather than fee simple titles to
meant that TVA was opposed to the trend of setting land aside under public owner-
ship ; or, as sometimes alternatively expressed, that TVA was interested in land in
relation to people, not in the land for its own sake.
17 The question of the practical desirability of such a course, or of the agricul-
turists' view, is not here in question. Important is that the controversy involved a
challenge to the older ideals accepted by the Authority, and consequently represented
pressure to effect a change in the basic character and meaning of the organization.
This is clarified even further by the agriculturists' assertion, often repeated, that
"there is really no such thing as sub-marginal land."
Selznick: TVA and the Grass Roots 203
tracts which would be flooded only occasionally. By late 1941 this prac-
tice was already being widely used. It is indicative of the general char-
acter and social implications of this trend that even where outright
public ownership had been foregone, and only flowage rights procured,
the opponents of public ownership pushed their demands so as to elimi-
nate the last elements of public control over land at the reservoir mar-
gins. In executing the policy of purchasing flowage easements, the
General Manager had authorized the Land Acquisition Department to
include a provision in all such contracts permitting public access to the
land covered by the easement for purposes of hunting, camping, fishing,
emergency landing, and the like. This would permit the reservation to
the public of recreational opportunities created by the new reservoir.
Under this policy, in case the Authority representatives were unable
to acquire flowage easements which included a public access provision,
outright purchase would be authorized. The inclusion of the public access
provision was considered especially important in such areas as that
around Fort Loudon Reservoir (near Knoxville) where a significant
movement of population to the reservoir shore line could be expected.
If such a movement occurred, and the right of public access from the
reservoir to the shore had not been reserved, it was anticipated that soon
the general public would be denied use of the lake created by the TVA
dam.
The general manager defended the public access provision in the
purchase of easements, though he suggested that the elimination of
camping as one of the reserved rights might quiet some objections. The
opponents of the public access provision within the Authority were
speaking for the local landowners ; in effect they were ready to relinquish
to a relatively few individuals the opportunities for use and enjoyment
created as a result of the technological changes introduced by TVA in
the area. The opposition to public ownership or control over land use
tended to wipe out that concern for the social consequences of its activi-
ties which represented an important aspect of the character of the
Authority as a whole. The public access provision remained intact, but
the significant successes of the agriculturists and their allies within
the Authority, in respect to the evolution of the land-purchase program
as a whole, seemed to weaken rather than strengthen the general man-
ager's position.
A significant victory for the agriculturists, as was generally recog-
nized within the Authority, came on February 2, 1942, when the Board
of Directors adopted a resolution defining a general policy on land
204 University of California Publications in Culture and Society
acquisition which definitively rejected the policy of acquiring a protec-
tive strip. The Board declared that it was now its policy "to limit the
purchase of land for reservoir purposes to the minimum appropriate
for the particular project. In general, land lying above the zone of
reservoir fluctuations" was not to be purchased in fee simple except
under certain specified conditions. In effect, the Board accepted the rea-
soning of those who opposed the protective strip and rejected that of its
proponents. The retreat from the early policy was now complete. From
the period of the construction of the Norris Dam, nine years earlier,
which envisioned and effected widespread public ownership and control
of rural lands, the policy had now become one of restricting public own-
ership to the minimum dictated by specifically authorized projects. The
economy orientation of the chief engineer and the agriculturists' repre-
sentation of local landowner interests seemed to have effectively altered
one of the basic aspects of the Authority's program. 18
The Board of Directors itself is said to have been split over the final
decision. It is reported that the resolution of February 2, 1942, was
passed in the absence of Chairman Lilienthal and that he was in fact
opposed to it. However, owing to another controversy then demanding
his attention, the chairman is said to have accepted the decision of the
other two members of the Board. It can scarcely be doubted that this
was in fact a victory for Harcourt A. Morgan and his group within the
Authority, for former Senator James P. Pope the third member of the
Board seems not to have exercised an independent role at this time.
It is probable that the abandonment of the protective strip policy was
aided by the public controversy in 1941 over the construction of Douglas
Dam. This project, located on the French Broad River near Dandridge,
Tennessee, was hotly opposed by representatives of the local area. The
leading antagonist was Senator Kenneth McKellar of Tennessee, and he
was supported by Governor Prentice Cooper, as well as representatives
of the local farm bureau organizations and President James D. Hoskins
of the University of Tennessee. 19 A long-standing argument against the
18 The agriculturists themselves were not overly concerned over program economies
as such; they were in favor of paying high costs for road relocation or severance
damages if that would permit land to remain in private ownership.
19 The public opposition of President Hoskins was a blow to the TVA, which had
long cooperated with the University of Tennessee, granting it administrative control
over personnel subsidized by TVA funds. In the Douglas Dam controversy, the
TVA's commitments to the defense program denned it as a representative of the
federal government in the Valley. To the extent that the Authority had to identify
itself with the federal program, it had to bear the brunt of opposition to that pro-
gram. TVA would have liked to have its cake and eat it too : contribute mightily to
the federal defense program and still retain the loyalties of those local institutions
and people who would be adversely affected.
Selznick : T VA and the Grass Roots 205
TV A that it brought more evil by flooding productive agricultural land
than the good it created was again given public attention. This must
certainly have strengthened the hand of the TVA agriculturists in their
fight for revision of the older policy.
However, it must be emphasized that the TVA agricultural group was
opposed not only to the acquisition of clearly valuable agricultural
land though acquisition did not necessarily mean such land would be
taken out of production but to all unflooded land, whatever its adapta-
bility for agricultural purposes. Thus they opposed the acquisition of
a large tract of forest land, which was to have been turned over to the
U. S. Forest Service as part of the national forest reserve, on the theory
that the Authority had no right to decide that the land should be publicly
owned forest. Moreover, having won their point on the protective strip,
the agriculturists did not rest content. They continued their opposition
to specific acquisitions, as for recreation purposes, and intensified their
demand that the Authority reduce its effect upon the agricultural econ-
omy to an absolute minimum. The latter objective was maintained even
when the cost of providing new means of access to some areas would
exceed the cost of acquiring the land, or when families would be left to
inadequate school services. In all such conflicts, the agriculturists in-
voked the authority of the local land-grant colleges whenever possible.
"FANATICISM" AND ADMINISTRATIVE DECISION
The consistent pressure brought to bear upon policy by the agriculturists
within the Authority was a source of considerable vexation to other
departments. This pressure occurred not only in respect to the reservoir
land-purchase policy but on tangential questions as well. We may cite
two such issues here, and then attempt to explain, in terms of the fore-
going analysis, a frequent comment on the behavior of the agriculturists
made by other groups inside TVA.
The close link between the TVA agriculturists and the extension serv-
ices sometimes resulted in embarrassment to the TVA Land Acquisition
Department. The latter was responsible for technically evaluating land
to be purchased, and for paying prices that would enable the landowners
to relocate without lowering their living standards. The Authority's dual
role of, on the one hand, attempting to make its purchases at reasonable
cost, and, on the other, of benefiting the farmer as much as possible,
might normally be expected to lead to difficulties and disagreements. But
these land-purchase activities seemed to find a special source of resist-
ance in the attitude of the agriculturists that other departments within
206 University of California Publications in Culture and Society
the Authority might not be much concerned about the welfare of farmers
in the Valley. Moreover, the extension service in the field, operating
wherever land acquisition was carried on, provided the TVA agricul-
turists with a means of implementing their own views.
The extension service officials in the counties could bolster their influ-
ence among the farmers by aiding them when condemnation proceedings
were instituted by the Authority. Thus one assistant county agent, whose
salary was indirectly paid by the Authority under the joint agricultural
program, testified against the Authority at a condemnation hearing.
Other extension officials carried on activities in support of the local land-
owners against the Authority. In addition, the extension officials found
ready allies among the TVA agriculturists, which enabled them to press
their demands through the TVA staff itself. Presumably, this furthered
the prestige of the extension service, since it afforded them an influence
which could be discreetly boasted of to the local farmers.
The TVA agriculturists executed their commitment to the policy of
the extension service with great consistency, and were highly conscious
of the need to develop a strategy within the Authority. Thus, in 1941,
extension service representatives requested that the TVA provide an
appraisal breakdown to farmers whose property was to be purchased.
The agriculturists accepted this policy as their own, and, after the
request was rejected by the Land Acquisition Department, 20 raised the
question among themselves whether it would be wise "to wage a fight
to force release of appraisal breakdown." Moreover, the TVA Director
of Land Acquisition did not hesitate to express his conviction that joint
criticisms by the TVA Department of Agricultural Relations and the
extension service organizations had been transmitted to the local land-
owners, and that some local organization had been encouraged whose
purpose was to put pressure upon TVA for an increase in land prices.
This situation reinforced and deepened antagonisms which tended to
isolate the agriculturists from other groups within the Authority.
Another situation which reflected the TVA agriculturists' role of rep-
resenting local landowners led to a virtual rupture of relationships, and
considerable bad feeling, between the Authority and another conserva-
tion agency, the Bureau of Biological Survey. 21 The Bureau was respon-
20 On the grounds that an appraisal breakdown furnished to property owners would
undermine the purchase program, since it would provide landowners with additional
grounds for contesting the Authority's prices ; i.e., appraisals of parts of a property,
rather than the property as a whole, would become subject to debate.
21 Transferred July 1, 1939, from the Department of Agriculture to the Depart-
ment of the Interior, later constituting part of Fish and Wildlife Service. The chief
of the Bureau, Dr. Ira N. Gabrielson, became chief of the Service.
Selznick: T VA and the Grass Roots 207
sible for the national wild-life conservation program, centering partly
around the establishment of wild-fowl and game refuges in various parts
of the country. Early contacts had been initiated by the TVA foresters
under E. C. M. Richards for the operation of a joint program within
the Tennessee Valley area and especially on TVA-owned lands. On July
7, 1938, a Wheeler Migratory Wildfowl Refuge was created by Execu-
tive Order of the President. The new refuge comprised an area of some
41,000 acres within the protective strip purchased by TVA at Wheeler
Reservoir near Decatur, Alabama. The Bureau of Biological Survey
was designated as the agency to administer the refuge, but the lands
remained under the primary jurisdiction of TVA. This meant that the
development of the area would be based largely upon cooperative agree-
ments and administrative relationships established and maintained by
the two agencies.
Contacts between the Bureau and TVA had been continuous since the
establishment of the latter agency in 1933, but from the viewpoint of
the Bureau these relationships had never been very satisfactory. In
addition to some broad criticisms of the nature of the TVA program, 22
there were two main sources of friction. One of these was the TVA
malaria control program, which involved the use of insecticides on a
large scale and the lowering of water levels to kill mosquito-breeding
vegetation. Both of these practices were considered by the Bureau to be
inimical to wildfowl subsistence needs, and the TVA was requested
to adjust its malaria control program so that the deleterious effect upon
wild fowl would be kept to a minimum. This difficulty, which persisted
over several years, appears to have resulted from conflicting technical
programs requiring administrative coordination, rather than of any dis-
agreements over principle. However, the second source of continuing
friction was somewhat different. The TVA followed a policy of leasing
its lands to local landowners for agricultural use, utilizing the extension
service organization and voluntary land-use associations to execute the
leasing program. This activity was carried on within the Wheeler Refuge
area, and as a consequence of what the Bureau considered bad agricul-
tural practices from the standpoint of conservation some rather bitter
criticism was launched against the Authority.
Relations between TVA and the Bureau became acute in 1939, largely
22 In 1934, the then chief of the Bureau criticized the TVA for failing to put its
wild-life program on an economic basis, saying that the TVA seemed to consider
that program "as important a unit in their plans as a caged canary bird hanging
in the dining room." This criticism seems to parallel the strictures voiced by TVA's
first chief forester.
208 University of California Publications in Culture and Society
as a result of the activities and viewpoint of the agricultural group
within the Authority. The TVA had made certain commitments to co-
operate with the Biological Survey with respect to land practices on
Wheeler Eefuge. The TVA agriculturists, however, apparently did not
feel that such commitments were binding without the approval of the
local extension service and county agricultural associations. 23 This had
certain significant consequences. In the first place, negotiations carried
on between the Bureau and TVA would involve an element of unreality
so long as it was assumed by one of the parties that there was still another
group, not formally represented, whose agreement would be needed
before commitments could be binding. Secondly, the Bureau was being
asked, in effect, to submit its technical judgments on wild-life conserva-
tion needs and practices to the arbitration of a group of local people
whose immediate interests were to be directly affected by the program.
The first of these consequences was probably the source of that difficulty
within TVA management which left it open to charges of bad faith by
the Bureau. 24 The second consequence of the introduction of the local
groups as a party jeopardized principles of public control which had
been accepted only after a long struggle by those interested in a national
wild-life conservation program.
The TVA agriculturists seem to have been bent on pursuing their
opposition to the reservation of land for nonagricultural uses. For his
part, the Chief of the Bureau of Biological Survey insisted that
... it is not our desire to obstruct the agricultural development of good land or to
interfere with the local economy by taking valuable agricultural land out of produc-
tion. However, certain practices which are now being carried on, such as plowing
to the water's edge, destructive and late burning, overgrazing, and disregard of
contour cultivation, are certainly not conservation measures in any sense of the word,
and it is my belief that the Authority is negligent in allowing such practices to
continue.
It is clear that the ultimate soil conservation objectives of the TVA
agriculturists were not in principle opposed to those of the Bureau. But
the TVA personnel insisted that progress along that line must be based
23 In effect, these county associations were creatures of the extension service, run
by the county agent's office. See pp. 231 ff.
24 These were strongly put, and repeatedly. On one occasion, in 1939, the chief
of the Bureau commented that "agreements, written or verbal, have been of little
consequence to the TVA." The seriousness of the breakdown of amicable relation-
ships between TVA and the Bureau was attested to by Chief Forester Baker in 1940.
He suggested that failure on the part of the Authority to resolve the critical situa-
tion might well lead to action on the part of the Bureau by that time known as
Fish and Wildlife Service which might bring about a Congressional investigation,
or some "similar action equally disastrous." The Chief Forester himself was appar-
ently in basic sympathy with the Service, but was helpless because of his relatively
weak organizational position within TVA.
Selznick : T VA and the Grass Roots 209
on the education of local people, through their own institutions ; on this
view, it would be better to have bad agricultural practices for a while
than to remove the land from use by local farmers. It is this difference
not so much of objective as of method which separated the agricul-
turists from the conservation tradition which the Bureau represented.
Basic assumptions as to method, however, such as the use of public own-
ership as an instrument of planning and control, become part of the
character of movements and organizations.
It was precisely within the sphere of this difference in method that
the chief incident which deepened the friction between the two agencies
occurred. Recognizing that the TVA had already made some commit-
ments to the local agricultural interests at Wheeler Refuge, it was agreed
that representatives of the Bureau would attempt to gain the coopera-
tion of those interests through the local county agents and leaders of
the county soil conservation associations. A meeting was held at Athens,
Alabama, on May 5, 1939, but, far from permitting the Bureau personnel
informal access to the local farm leaders, an unexpectedly formal gather-
ing took place, attended by the assistant director of the TVA Agricul-
tural Relations Department, among others. In a general discussion a
TVA representative reportedly spoke disparagingly of the program of
the Biological Survey, and suggested to the local farm representatives
present that the establishment of the refuge was not necessarily perma-
nent and might be revoked if the members of the local soil conservation
association so desired. The public character of this meeting lent a special
importance to the remarks of the TVA official, and the Bureau was quick
to conclude that members of the TVA staff were deliberately attempting
to publicly discredit the Bureau and to undermine its program. The
consequences of such activity for amicable relations among public agen-
cies need hardly be labored.
The record of this controversy emphasizes once more the ambivalence
of the grass-roots approach. On the one hand, TVA's reliance upon the
extension service machinery permits its program to be shaped by local
interests and local demands. At the same time, the TVA as a whole is
forced into embarrassment and maladjustment with respect to the regu-
lar federal organization. Normally any regional agency would be torn
between its desire to adapt itself to local interests and at the same time
to execute a unified national policy. A series of compromises would be
expected. This is likely to occur even in an organization formally cen-
tralized, with headquarters in Washington. But TVA has gone farther.
In delegating its agricultural program to a group which carries the
210 University of California Publications in Culture and Society
banner of a local constituency, the Authority comes to reject in prin-
ciple not simply as a matter of temporary expediency those elements
of the national program which are opposed by the constituency.
In commenting upon the persistent pressure of the agriculturists to
shape the policy of TVA on land purchase and land use, other groups
within the Authority have characterized their behavior as fanatical.
Not uncommon are exasperated references to the apparent willfulness
of the agricultural group, to their easy lapse into "preachy" discussion,
to the fact that they often constitute a "minority of one" 25 among the
interested departments. These are extreme characterizations, and are
not necessarily based upon a sober or rounded appraisal of the role
of the Agricultural Relations Department. However, one may ask:
what organizational significance can we infer from this impression of
"fanatical" behavior ?
The behavior of a faction within an organization assumes a quality
of strangeness when it appears to be reckless of the basic unity of the
whole group. 26 It is the continuity of basic assumption which knits
together the several administrative units : it is that continuity which
is presumed if methods of formal coordination are meaningful. When,
however, a group is divided on matters of fundamental significance,
the real locus of decision will tend to move out of the sphere of formal
administrative channels. The breakdown of real unity will be reflected
in administrative coordination becoming no longer a mechanism for
the resolution of technical and transitory maladjustments but rather
a vehicle and an instrument in the unacknowledged struggle over a
primary prize : the evolving character of the organization as a whole. 27
Thus an examination of the context of the TVA Budget Office move,
in 1943, to tighten control over the Agricultural Relations Depart-
ment 28 reveals that an important motivating factor arose from the
doubts felt by the Budget Office personnel that the agriculturists were
25 Which need not mean defeat, since decisions are of course not made by taking
a vote of staff members or sections.
26 This whole discussion of "fanaticism" is informed by Max Weber's distinction
between Gesinnungsethik ("ethic of ultimate ends") and Verantwortungsethik
("ethic of responsibility"). The peculiar role of the agriculturists becomes quite
understandable when it is seen as ordered by ends which are ultimate or absolute
with respect to the organizational system in which they are participants. See H. H.
Gerth and C. Wright Mills, From Max Weber (New York : Oxford University Press,
1946), p. 120 f.
27 This principle is more obvious in, but by no means restricted to, formally demo-
cratic organizations, as when, in periods of dispute, the meetings of an executive
committee may serve only to preserve the fiction of unity whereas the real decisions
are being made on a factional basis, in camera.
28 See above, p. 115.
Selznick : T VA and the Grass Roots 211
in fact carrying out a program consistent with the ideals which brought
the Authority into being. In general, it was possible to gather a sense
of uneasiness with respect to the agriculturists a feeling that some-
how this particular group was not an integrated part of the TVA as a
whole and did not completely share its aspirations. This feeling was by
no means universal, but sufficiently widespread to offer a basis for
comment and reflection.
The characterization of the agriculturists as fanatic doubtless finds
its roots in this sense of underlying disunity. For it is this disunity
which makes possible action which, from the point of view of the group
as a whole, is irresponsible. The quality of responsibility in action
arises when the basic commitment of an individual or a faction is to
the integrity of the whole ; it involves a willingness to compromise, to
accommodate special outlooks and goals, and presumes an acceptance
of common principles. When, however, there is a sense of isolation, of
alienation, action may be shaped by criteria of conflict and disaffection.
Moreover, when the cleavage is not fully recognized, other groups may
simply feel that "something is wrong" and offer psychological explana-
tions for what appears to be strange and discordant behavior.
With these considerations in mind, it is possible to suggest three
factors which, taken together, may offer a suggestive explanation of
the label "fanatic" which has been applied, in some quarters of the
Authority, to its agricultural organization.
1. The agriculturists are the only group in the TVA which, in a
unified way, is representative of the local area. 29 The personnel of the
Agricultural Relations Department, and its leadership in particular,
come not only from the area but from a special institution : the agricul-
tural extension services. They share many of the basic attitudes indig-
enous to the area, 30 including hostility to those who come from afar
to effect changes and offer criticism based on alien values. Consequently,
it seems fair to suggest that the agriculturists must look upon the rest
of the Authority as something imposed upon the local area, whose
capacity for evil is great and must be limited as much as possible. It is
difficult to avoid the conclusion based upon the arguments offered by
the agriculturists in the land-purchase policy controversy that it is
only expediency which has restrained them from supporting those 31
29 See above, p. Ill f.
30 "As between pellagra and tourists," one of the leading agriculturists is reported
to have said, "I'll take pellagra."
31 Such as Andrew J. May of Kentucky, formerly chairman of. the House Military
Affairs Committee.
212 University of California Publications in Culture and Society
who have expressed the belief that the damage done by the Authority
in flooding valuable agricultural lands is greater than the good accom-
plished. It is probable that the sense of being representative gives the
agriculturists the feeling that they are playing a historic role in the
interests of their own people and against the outsiders.
2. Evidence has been presented above that a constituency relationship
was constructed involving the TVA agriculturists on the one side and
the land-grant college system on the other. On this theory, much of the
behavior of the agriculturists may be viewed as a function of external
commitment. This serves as another factor in explaining behavior which
appears to be irresponsible. Given an external commitment, responsi-
bility becomes divided or even completely referrable to the loyalties
of the subgroup to its constituency. Such terms as "fanatical" are em-
ployed when the real basis of irresponsibility is not clearly understood.
The "fanaticism" of American Communists, and their putative irre-
sponsibility, becomes clear when the commitment of the Communist
Party to the regime in the USSR is understood. What is irresponsible
from the point of view of American national interests becomes clearly
responsible from the point of view of the Soviet Union. Less dramat-
ically, but by the same logic, the actions of the agriculturists can be
fully understood only in relation to their constituency commitments.
3. "Fanaticism" may also be a label for that type of behavior which
bases itself upon the persistent espousal of a general doctrine or for-
mula, in such a way that issues are not considered upon their merits.
This is a very familiar use of the label in secular contexts. If concrete
problems are simply occasions for the expression of a general point
of view, if the special goal of a particular individual or group is always
put forward without regard to what most participants consider an
appropriate occasion, the use of the term "fanatic" is normally in order.
But the characterization may often have little to do with the person-
ality of the individual involved. It is his method which counts, a method
which may be consciously adopted under organizational pressures and
in fact inimical to his own psychology. Thus a man with a special doc-
trinal axe to grind, as, say, that the interests of workers and employers
are antithetical, may squeeze doctrinal implications out of each con-
crete issue, and take his position on the level of doctrine. In the same
way, the TVA agriculturists have had certain general goals in mind
the defense and strengthening of the land-grant college system and the
limitation of public ownership of land and they have handled so
many concrete issues according to their implications for these goals
Selznick : T VA and the Grass Roots 213
that others in the Authority have come to think of them as having an
unreasonable approach.
Hence we can see the organizational basis for the so-called fanaticism
of the agriculturists : alienation, induced by the position of the group
relative to others; action as a function of external commitment; and
the persistence of intervention on the level of doctrine. At the same
time, it is clear that such an explanation could not have been reached
apart from the analysis of the concrete meaning of the grass-roots
approach.
Part Three
THE FORMAL COOPTATION OP
GRASS-ROOTS ELEMENTS-
AMBIGUITIES OF PARTICIPATION
CHAPTER VII
THE VOLUNTARY ASSOCIATION AT THE
END POINT OF ADMINISTRATION
In our state, we feel that the extension service must control the system of community
organizations, county associations, and neighborhood leaders. Other agencies must
come through the extension service. After all, it's our baby.
A SUPERVISOR ON THE JOINT TVA-ExTENSiON SERVICE PROGRAM (1943)
THE CONSTRUCTION of an administrative constituency, whereby the
dominant agricultural leadership in the Tennessee Valley area was
afforded a place within the policy-determining structure of the TV A,
is an example of the process of informal cooptation. The preceding
chapters have traced the origin and structural details of the constitu-
ency relationship and have sought out its unanticipated consequences
for the role and character of the Authority. The explication of these
consequences functions at once as evidence for, and as a key to the
meaning of, the cooptative mechanism. The unacknowledged absorp-
tion of nucleuses of power into the administrative structure of an
organization makes possible the elimination or appeasement of poten-
tial sources of opposition. At the same time, as the price of accommoda-
tion, the organization commits itself to avenues of activity and lines
of policy enforced by the character of the coopted elements. Moreover,
though cooptation may occur with respect to only a fraction of the
organization, there will be pressure for the organization as a whole to
adapt itself to the needs of the informal relationship. Viewed thus
broadly, the process of informal cooptation represents a mechanism of
comprehensive adjustment, permitting a formal organization to en-
hance its chances for survival by accommodating itself to existing cen-
ters of interest and power within its area of operation.
This analysis directs attention to the unavowed meaning of the official
grass-roots policy. One of the major tenets of that policy is the injunc-
tion that the program of a regional agency be channeled through the
existing institutions of the area of operation, and that a positive policy
of strengthening those institutions be maintained. In formal terms,
this is taken to mean that the institutional resources of the local area
will be fully utilized, and a democratic partnership with the people's
institutions effected. The informal organizational consequences of that
injunction depend on the relative strength and the respective prob-
[217]
218 University of California Publications in Culture and Society
lems of the organizations involved, and are not readily apparent before
analysis. It is that analysis which the preceding chapters have attempted
to provide.
But there is another phase of the grass-roots policy which deserves
similar attention. The concept of a regional partnership includes the
idea that ordinary citizens will be drawn into the administration of
the regional program through membership in voluntary associations. 1
Wherever the execution of the over-all program reaches out into a
local community, it is considered desirable to organize those citizens
most closely affected into an association which will participate in the
administration of the program. There is thus envisioned a mushroom-
ing of citizens' organizations at the end point of administration, per-
mitting participation at the grass roots in the application of a general
policy to varying local conditions. The TVA has raised this procedure
to an administrative principle, treating it as a fundamental aspect of
what it considers to be a unique grass-roots approach.
The policy of encouraging citizen participation through the creation
of voluntary associations should be clearly differentiated, in terms of
its implications for democratic administration, from that approach
which assumes the legitimacy of certain people's institutions, through
which action at the grass roots must be channeled. The use of voluntary
associations does not of itself determine whether the existing institu-
tions of an area are to be utilized. Indeed, the construction of voluntary
associations ad hoc may serve as a substitute for the channeling
of a program through established organizations. Associations created
through the direct intervention of a federal agency, to implement the
administration of its own program, may result in the establishment
of new channels to the population, by-passing an existing leadership
or established institutions. This possibility is at hand whenever an
appeal is made to the individual citizen rather than to a previously
instituted local leadership. Therefore it is not always clear that these
two separate tenets of the grass-roots approach reinforce or even neces-
sarily complement each other. In creating the electric power coopera-
tives, the Authority reached out directly to organize the local citizens
for participation in the execution of the program of the regional agency.
This represented the creation of new centers of organizational strength
without the mediation of any previously existing institutions. 2 In
the agricultural program, the Authority has made working with and
1 See above, p. 40.
2 Yet this type of direct action on the part of other federal agencies, working out
of Washington, has been roundly criticized by TVA officials.
Selznick : T VA and the Grass Roots 219
through the extension service organizations its primary grass-roots
policy. Thus even within TVA experience, the use of voluntary associ-
ations may or may not be connected with the policy of siphoning the
program through established institutions. In general, it appears that
specific programmatic and administrative imperatives, rather than gen-
eral considerations of democratic policy, determine the form of inter-
vention at the grass roots.
It follows that the significance of two tenets of the grass-roots
theory working through existing institutions and using voluntary
associations must differ considerably. "We have already interpreted
the one as informal cooptation, the adjustment to concrete forces
within the Valley. In approaching the other, we shall utilize the con-
cept of formal cooptation, in which is uppermost the need to share the
responsibilities for or the administrative burdens of power, rather
than power itself. 3
FORMAL COOPTATION AND AGRICULTURAL
DEMOCRACY
The use of voluntary associations is not new, and is far from unique
or peculiar to the program or administration of the Tennessee Valley
Authority/ Indeed, it is useful to think of the cooptation of citizens
into an administrative apparatus as a general response made by gov-
ernments to what has been called "the fundamental democratization"
of society. 5 The rise of the mass man, 6 or at least the increasing need
for governments to take into account and attempt to manipulate the
sentiments of the common man, has resulted in the development of new
methods of control. These new methods center about attempts to organ-
ize the mass, to change an undifferentiated and unreliable citizenry into
a structured, readily accessible public. Accessibility for administrative
3 See above, pp. 13-14, and below, pp. 259-261.
* Perhaps as testimony to the effectiveness with which the grass-roots doctrine is
circulated within the TVA, there appears to have developed a feeling that the Author-
ity has somehow originated a unique administrative device, binding the agency to
its client public in some special way. This is partly referrable to enthusiasm, partly
to the prevalent idea that other federal agencies, lacking the halo of regional de-
centralization, are unlikely to be really interested in democratic administration.
It is hardly necessary to enter that controversy here, or to lay undue emphasis upon it.
Yet although the grass-roots method is considered one of the major collateral objec-
tives of the Authority, relatively little attention has been paid to the mechanics of
its implementation and certainly the experience of other organizations facing the
same problems and using voluntary associations has not been seriously studied inside
the Authority.
5 Karl Mannheim, Man and Society in an Age of Reconstruction (New York: Har-
court, Brace, 1941), pp. 44 ff.
6 See Jose Ortega y Gasset, The Revolt of ihe Masses (New York: Norton, 1932).
220 University of California Publications in Culture and Society
purposes seems to lead rather easily to control for the same or broader
purposes. Consequently, there seems to be a continuum between the
voluntary associations set up by the democratic (mass) state such as
committees of farmers to boost or control agricultural production and
the citizens' associations of the totalitarian (mass) state. Indeed, the
devices of corporatism emerge as relatively effective responses to the
need to deal with the mass, and in time of war the administrative tech-
niques of avowedly democratic countries and avowedly totalitarian
countries tend to converge.
Democracy in administration rests upon the idea of broadening par-
ticipation. Let the citizen take a hand in the working of his govern-
ment, give him a chance to help administer the programs of the positive
state. At its extreme, this concept of democracy comes to be applied to
such structures as conscript armies, which are thought to be democratic
if they include all classes of the population on an equal basis. If analysis
and appraisal is to be significant, however, it is necessary to inquire
into the concrete meaning of such an unanalyzed abstraction as "par-
ticipation." In doing so, we shall have to distinguish between substan-
tive participation, involving an actual role in the determination of
policy, and mere administrative involvement. In the conscript army,
we have a broadening involvement of citizens, with a concomitant ab-
dication of power. The same may be said of the Japanese tonari gumi,
neighborhood associations which helped to administer rationing and
other wartime programs. Such organizations, which have had their
counterparts in many parts of the world, involve the local citizens, but
primarily for the convenience of the administration. It is easy enough
for administrative imperatives which call for decentralization to be
given a halo; that becomes especially useful in countries which prize
the symbols of democracy. But a critical analysis cannot overlook that
pattern which simply transforms an unorganized citizenry into a re-
liable instrument for the achievement of administrative goals, and calls
it "democracy." 7
The tendency for participation to become equivalent to involvement
has a strong rationale. In many cases, perhaps in most, the initiation
of local citizens' associations comes from the top, and is tied to the
pressing problem of administering a program. The need for uniformity
7 This is no necessary reflection on the integrity or the intentions of the responsible
leadership. It is normal for programs infused with a moral content to be reduced
to those elements in the program which are relevant to action. Thus the moral ideal
of socialism has been reduced rather easily to concrete objectives, such as national-
ization of industry. Administrative objectives, such as the establishment of a rami-
fied system of citizens' committees, are similar.
Selznick : T VA and the Grass Roots 221
in structure; for a channel through which directives, information, and
reports will be readily disseminated ; for the stimulation of a normally
apathetic clientele ; and for the swift dispatch of accumulated tasks
these and other imperatives are met with reasonable adequacy when
involvement alone has been achieved. Some additional impetus, not
provided for in the usual responsibilities of the administrative agency,
is normally required if the process is to be pushed beyond the level of
involvement. Indeed, it is doubtful that much can be achieved beyond
that level. Such associations, voluntary or compulsory, 8 are commonly
established ad hoc, sponsored by some particular agency. 9 That agency
is charged with a set of program responsibilities. These cannot be read-
ily changed, nor can they be effectively delegated. As an administrative
organization, the agency cannot abandon the necessity for unity of
command and continuity of policy not only over time but down the
hierarchy as well. What, therefore, can be the role of the coopted local
association or committee ? It cannot become an effective part of the major
policy-determining structure of the agency. 10 In practice only a limited
sphere of decision is permitted, involving some adaptation of general
directives to local conditions, and within that circumscribed sphere
the responsible (usually paid) officials of the agency will play an effec-
tive part.
With these general considerations in mind, it may be well to mention
at least one phase of the historical context within which the TVA's use
of voluntary associations has developed. Especially in the field of agri-
cultural administration, the TVA's methods have paralleled an emerg-
ing trend in the administration of the federal government. This is not
often recognized within the Authority, but there can be little doubt
that the United States Department of Agriculture has gone much far-
ther in developing both the theory and the practice of citizen participa-
tion than has the TVA. The emergence of this trend accompanied the
construction of a vast apparatus to administer an action program reach-
ing virtually every farmer in the nation.
8 This distinction tends to melt away as the program administered comes closer
to becoming an exclusive means of distributing the necessities of life, or if induce-
ments are such as to eliminate any practical alternatives to participation.
9 This may be the effective situation, even where there is not legal sponsorship.
Thus, the local soil conservation districts are creatures of the state legislatures, but
serviced by the Department of Agriculture's Soil Conservation Service. It is prob-
ably not inappropriately that they have been known in some areas as "SOS" districts.
^ 10 This might happen if the local groups formed an independent central organiza-
tion, but that is not envisioned by the administrative agency, unless it already has
control of a preexisting central organization, as when a national government utilizes
a preexisting party structure to aid in the administration of its program.
222 University of California Publications in Culture and Society
One formulation of the idea of "agricultural democracy" was under-
taken in 1940 by M. L. Wilson, Director of Extension Work of the
Department of Agriculture. 11 Wilson noted the movement toward a
greater group interest on the part of farmers, and the pressure for
equality through government intervention, culminating in the enact-
ment of the Agricultural Adjustment Act of 1933 and subsequent New
Deal agricultural legislation. The administration of the new govern-
men programs was based on the ideal of cooperation and voluntary
participation, leading to a set of procedures which, in Wilson's view,
can be thought of as the general principles of agricultural democracy :
1. Decentralized administration in varying degrees through community, county,
and state farmer committees, elected by cooperating farmers or appointed by the
Secretary of Agriculture.
2. The use of referendums in determining certain administrative policies, es-
pecially those having to do with quotas, penalties, and marketing agreements.
3. The use of group discussion and other adult education techniques as a means
of promoting understanding of the problems and procedures involved in admin-
istration of the various programs and referendums.
4. Cooperative planning in program formulation and localization of programs. 12
This program emphasizes the importance of participation within the
democratic pattern of culture. Moreover, in theory, participation in-
cludes both policy-forming and administrative functions.
The technique of coopting local citizens through voluntary associa-
tions and as individuals into the administration of various agricultural
programs was widely developed during the nineteen-thirties. In 1940,
it was reported that over 890,000 citizens were helping to plan and
operate nine rural action programs: 13 community, county, and state
committees of the Agricultural Adjustment Administration, operating
through over 3,000 county agricultural associations ; land-use planning
committees organized through the Bureau of Agricultural Economics ;
farmer associations aiding in the administration of Farm Credit
Administration loans; rehabilitation and tenant-purchase committees
11 "A Theory of Agricultural Democracy," an address before the American Political
Science Association, Chicago, December 28, 1940. Published as Extension Service
Circular 355, March, 1941 (mim.). See also M. L. Wilson, Democracy Has Eoots (New
York: Carrick & Evans, 1939), chap. vii. Also, Howard E. Tolley, The Farmer Citizen
at War (New York: The Macmillan Co., 1943), Pt. V.
12 Wilson, "A Theory of Agricultural Democracy," p. 5. It is interesting to note
that Mr. Wilson, Director of Extension Work, considered the AAA program to have
represented the practical beginning of agricultural democracy. The TVA agricul-
turists, loyal essentially to the local extension service organizations, would not have
made such a statement.
13 Carleton K. Ball, "Citizens Help Plan and Operate Action Programs," Land
Policy Eeview (March-April, 1940), p. 19.
Selznick : T VA and the Grass Roots 223
organized by the Farm Security Administration ; local district advisory
boards for the Grazing Service; cooperatives dealing with the Rural
Electrification Administration ; governing boards of soil conservation
districts serviced by Soil Conservation Service; these together form
an administrative pattern of which the TVA ventures along this line
were only a part. Mr. Ball's summary of participating citizens is re-
printed as table 9.
The trend toward cooptation of farmers in the administration of a
national agricultural program reached a high point with the organiza-
TABLE 9
SUMMARY OF ASSISTING CITIZENS IN U. S. AGRICULTURAL PROGRAMS 1939
Name of citizen group Members
AAA: local committees 135,591
County land-use committees 72, 000
Extension Service: volunteer program leaders 586,600
FCA: association directors and committees 36,574
FSA: local committees 26,753
Grazing Service: district advisory boards 547
REA: association directors 4, 900
Soil conservation district supervisors 855
Tennessee Valley committees and test-demonstrators 29,035
892,855
SOURCE: Carleton R. Ball, "Citizens Help Plan and Operate Action Programs," Land Policy
Review (March-April, 1940), p. 26. In 1942, just prior to the withdrawal of federal support
from the program, membership in state, county, and community land-use planning committees
reached 125,000, according to the Report of the Chief of the BAE, 1942.
tion of the county land-use planning program in 1938. The idea of
democratic planning with farmer participation was given considerable
attention, and an attempt was made to construct a hierarchy of repre-
sentative committees which would embody the democratic ideal. 1 * At
the same time, the achievement of a primary administrative objective
was envisioned. The land-use planning organization program received
its impetus from a conference of representatives of the Department of
Agriculture and the Association of Land-Grant Colleges and Univer-
sities held at Mt. Weather, Virginia, in July, 1938. The Mt. Weather
14 See "The Land Use Planning Organization," County Planning Series No. 3,
Bureau of Agricultural Economics, May, 1940 ; also, Land Use Planning Under Way,
prepared by the BAE in cooperation with the Extension Service, FSA, SCA, AAA,
and Forest Service, USD A, Washington, July, 1940 ; and John M. Gaus and Leon O.
Wolcott, Public Administration and the U. S. Department of Agriculture (Chicago:
Public Administration Service, 1940), pp. 151 ff.
224 University of California Publications in Culture and Society
Agreement 15 recommended a system of coordinated land-use planning
to overcome the confusion created by the existence of a large number
of points of contact between governmental agencies and local farmers.
By providing that local officials of the national agricultural agencies
would be represented on the farmer committees, it was felt that a
single point of contact would be established. It is possible that the
land-use planning system would have been established without this
impetus from a pressing administrative imperative, but it is clear that
the latter was the occasion for the new organization. The problems
of the officialdom were primary, and logically so, for their responsi-
bilities had to do with the efficient execution of statutory programs
not the creation of new culture patterns. The latter might, time and
resources permitting, have become an effective collateral objective, but
it would be idle to suppose that the requirements of administration
would not assume priority within the system.
The cooptative construction of systems of voluntary associations ful-
fills important administrative needs. These are general, and include:
1. The achievement of ready accessibility, which requires the estab-
lishment of routine and reliable channels through which information,
aid, and requests may be brought to segments of the population. The
committee device permits the assembling of leading elements on a reg-
ular basis, so that top levels of administration may have reason to
anticipate that quota assignments will be fulfilled; and the local
organization provides an administrative focus in terms of which the
various line divisions may be coordinated in the field.
2. As the program increases in intensity it becomes necessary for the
lower end of administration to be some sort of group rather than the
individual citizen. A group-oriented local official may reach a far larger
number of people by working through community and county organ-
izations than by attempting to approach his constituency as individuals.
Thus the voluntary association permits the official to make use of un-
tapped administrative resources.
3. Administration may be decentralized so that the execution of a
broad policy is adapted to local conditions by utilizing the special knowl-
edge of local citizens ; it is not normally anticipated, however, that the
policy itself will be placed in jeopardy.
15 Keprinted as an appendix to Gaus and Wolcott, op. cit. Russell Lord (The
Agrarian Revival, p. 193) notes extension service references to this agreement as
"The Truce of Mt. Weather." Truce indeed, for by the middle of 1942 the Congress
had scuttled the program, with the support of the American Farm Bureau Federa-
tion. See Charles M. Hardin, "The Bureau of Agricultural Economics Under Fire:
A Study in Valuation Conflicts," Journal of Farm Economics, Vol. 28 (August, 1946).
Selznick : T VA and the Grass Roots 225
4. The sharing of responsibility, so that local citizens, through the
voluntary associations or committees, may become identified with and
committed to the program and, ideally, to the apparatus of the oper-
ating agency.
These needs define the relevance of the voluntary-association device
to the organizational problems of those who make use of it It is only
as fulfilling such needs as these that the continuity in both structure
and function of this type of cooptation under democratic and totali-
tarian sponsorship can be understood.
From the above it is not surprising that criticisms of the county plan-
ning program have stressed deviations from the democratic ideal,
particularly in lack of representativeness and the tendency for estab-
lished organizations such as the Farm Bureau to take control of the
local committees. 16 Insofar, however, as this represents criticism of a
program developing toward complete fulfillment of the ideal, it is not
basic. More significant for this analysis are such criticisms as the
following :
... it is the central thesis of this paper that county planning did not succeed be-
cause no desire to solve community and county problems was created in the popu-
lation of the area in which the county planning program was to function Most
administrators of county planning conceive of rural planning as another admin-
istrative problem, as a procedure."
The normal pattern perhaps inevitable because of the rapid creation
of a nationally ramified system of committees established an organiza-
tion set down from above, oriented toward the administration of the
national program. As a consequence, the problems of the local official
qua official assumed priority. "One needed only to talk with represent-
atives of the several agencies engaged in trying to 'enforce' the county
planning system to recognize how ubiquitous this condition [of apathy]
was." u To the extent that the problems of the officialdom are sufficiently
pressing to stamp the character of the organization, it may be expected
that involvement rather than meaningful participation will prevail.
This same point is made in another way by John D. Lewis, in tracing
one of the bases for the lack of complete representativeness.
16 See John D. Lewis, "Democratic Planning in Agriculture," American Political
Science Review, XXXV (April and June, 1941) ; also Neal C. Gross, "A Post Mortem
on County Planning," Journal of Farm Economics, XXV (August, 1943) ; and Bryce
Kyan, "Democratic Telesis and County Agricultural Planning," Journal of Farm
Economics, XXII (November, 1940).
17 Gross, p. 647.
18 1~bid., p. 653. Mr. Gross also points out that the units of planning tended to fol-
low the convenience of the administrators, rather than local interest patterns.
226 University of California Publications in Culture and Society
The pressure to "get things done" has tended to encourage appointment rather than
election. The Division in Washington naturally expects its field agents to report
results that will justify the high hope with which the program was launched, and
the state office in turn pushes the county agents for progress reports with which
to appease Washington administrators. Democratic procedure is notoriously slow
procedure. Consequently the first thought of an overworked county agent, unless he
is genuinely impressed by the importance of finding a truly democratic committee,
will be to find a group of industrious and cooperative farmers who can be depended
upon to work together harmoniously. With the best of intentions and with no
thought of deliberately stacking the committee, he may set up a committee of "out-
standing leaders" who have a sincere desire to act in the interest of the whole
county, but who have only a second-hand knowledge and indirect concern about
the problems of less successful farmers in the county. 19
In effect, those responsible for organizing the system of committees or
associations are under pressure to shape their actions according to
exigencies of the moment, and those exigencies have to do primarily
with the needs of administration. As the needs of administration be-
come dominant, the tendency for democratic participation to be reduced
to mere involvement may be expected to increase. At the extreme, the
democratic element drops out and the cooptative character of the
organizational devices employed becomes identified with their entire
meaning. 20
EVOLUTION OF TVA INTEREST IN COOPERATIVES
The pattern of formal cob'ptation of local citizens has been, as we have
seen, rather widely developed in connection with major programs of
the federal government. Recognition of this permits us to examine
similar developments undertaken by TVA without ascribing to them
a uniqueness which they do not possess. By the same token, an analysis
of the general pattern may be expected to retain its significance when
applied to the TVA ventures.
The existence of formal cooptation cannot be inferred directly from
the bare form of any given organizational device, for it is its function
in fulfilling organizational needs which is decisive. Thus one cannot
conclude that merely because the TVA has been interested in the de-
velopment of local cooperative enterprises that these have functioned,
or were intended to function, as a means of mustering popular sup-
19 Lewis, op. tit., p. 247.
20 Unless "democracy" is reinterpreted, so that it reaches a higher level with the
subordination of the mass to the organization. The above account, one sided in its
emphasis, in no way deprecates the democratic aims of the initiators of the planning
program. We are concerned here only with the explication of underlying trends to
which the concept of cooptation lends significance.
Selznick: T VA and the Grass Roots 227
port for the agency. Some TVA officials have indeed expressed the
opinion that the voluntary associations developed in connection with
Authority programs do operate to create a mass base for the agency.
But the available evidence does not seem to warrant any general con-
clusion that there has been organized a manipulable source of public
support. However, insofar as devices have been created which permit
the sharing of responsibility for specific aspects of the program, as
well as the solution of administrative problems, it does make sense to
interpret the use of voluntary associations by the TVA as formal
cooptation.
The TVA's interest in the organization of local cooperative associa-
tions began with an emphasis on the development of cooperatives as
such; cooperation as a form of economic enterprise was furthered in
the interest of the general welfare of the region. However, as the pro-
grammatic orientations of the Authority came to be more clearly de-
fined and as certain earlier emphases were stripped away the interest
in local voluntary associations increasingly stressed the implementation
of specific subject-matter obligations. In this way the voluntary associa-
tion device came to be closely tied to the substantive responsibilities
and administrative needs of the agency. Concern for the creation and
strengthening of cooperatives was not abandoned. But this concern was
shaped, and to some extent inhibited, by the problems and emphases
of the major programs in connection with which the voluntary associa-
tions were organized. The history and rationale of this development
may now be reviewed, with particular reference to organizational issues.
Early interest in the furtherance of cooperatives as part of the
general- welfare objectives of the Authority seems to have come pri-
marily from Chairman Arthur E. Morgan. 21 He had long been inter-
ested in the development of local self-help programs and was quick to
turn the attention of the Authority in this direction. The opportunity
for initiating such a program came from a rather devious source, and
resulted in some confusion, both administrative and legal. Late in 1933,
negotiations were initiated between the Authority and the Federal
Emergency Belief Administration to have the latter agency allocate
a portion of its funds to a Tennessee Valley project for the organiza-
tion and financing of cooperatives. Presumably it was felt that the
objectives of both agencies would be served, if the funds were utilized
to help curtail immediate demands for relief, and, at the same time,
21 Organization of voluntary associations connected with electric power and ferti-
lizer distribution also began early in the history of the Authority, but primarily as
devices for the implementation of the major programs.
228 University of California Publications in Culture and Society
to provide a demonstration of methods which could help to raise the
level of living in the area.
As a consequence of these negotiations, a corporation was incorpo-
rated on January 25, 1934, under the laws of the State of Tennessee,
as the Tennessee Valley Associated Cooperatives, Inc., to receive and
administer the FERA grant. The incorporators were David E. Lilien-
thal, Har court A. Morgan, and Arthur E. Morgan, then directors of
the TVA, acting in their private capacity as individuals. By March
of that year an outright grant of $300,000 had been received by TVAC
from FERA, and in due course an administrator was appointed. Until
1936, the administration of the projects organized through TVAC rep-
resented the sum of TVA activities in this field. No additional funds
were authorized, but during this period TVA assumed responsibility
for the administrative expenses of the TVAC.
The work of TVAC consisted of financial assistance and advisory
service extended to cooperatives organized in certain depressed areas
of the Valley, especially in western North Carolina and eastern Ten-
nessee. Two groups of cooperatives received primary attention. One,
located primarily in the mountainous areas of western North Carolina,
consisted of cooperatives organized to market fruit and vegetable pro-
duce and operate a number of small canning plants. A regional associa-
tion to serve these cooperatives was established, known as Land o' the
Sky Mutual Association, through which the Authority personnel aided
the local groups. The other major project resulted from a study con-
ducted by the Women's Bureau of the United States Department of
Labor, which had investigated conditions under which a large number
of persons living in the Southern Appalachian Mountain Region were
producing and marketing handicraft articles as an important source
of family income. It was believed that the low cash income of these
people might be raised through a cooperative marketing project, and
so a regional organization was formed known as Southern Highlanders,
Inc., through which financial and advisory service was extended.
In addition, a number of miscellaneous cooperatives received loans
from TVAC, and some experiments in new processing ventures were
undertaken.
In 1935, a Cooperative Research and Experiment Division was estab-
lished within the Authority, primarily to administer the work of TVAC.
The authority for the work of the Division was based on Section 22 of
the TVA Act of 1933, which permits "studies and experiments ... for
the general purpose of fostering an orderly and proper physical, eco-
SelznicJc : T VA and the Grass Roots 229
nomic and social development" of the area. 22 This meant that the work,
in order to maintain its legal foundation, had to be framed in terms
of research and demonstration. The Authority did not have power to
accept the value of cooperatives as given and then conduct aggressive
promotional activities. The weakness of this delegation of authority in
the organic Act underlies much of the ambiguity of TVA developmental
programs.
After some disquiet as to the tenability of the relation between TVA
and TVAC, and perhaps as a minor by-product of the split within the
TVA Board of Directors, it was decided to separate the two organiza-
tions. In March of 1936, two of the three members of the TVA Board,
Harcourt A. Morgan and David B. Lilienthal, resigned as directors
of TVAC. They were replaced by Ray Crow, head of the Alabama
Relief Administration, and Clarence Pickett, advisor to the Resettle-
ment Administration. Arthur E. Morgan retained his place on the
TVAC Board and continued in this capacity after leaving the Author-
ity. In December, 1937, administrative relations between the Authority
and the TVAC were terminated, although some payments under con-
tracts with Southern Highlanders and Land o' the Sky continued
into 1938.
The separation from Tennessee Valley Associated Cooperatives
marked the end of the period wherein self-help cooperatives were pro-
moted or aided for themselves, apart from specific program objectives
of the Authority. Some doubt has been expressed within the agency
as to the value of the work which was carried on, and especially as
to whether the limited range of activities made possible any judgment
as to the appropriateness of cooperative enterprise in the Valley. Of
course, such a question can be seriously posed only if it is considered
to have been the real objective of the program to carry on "studies,
experiments, and demonstrations" as provided for in the Act. In fact,
the organizers of the program were probably quite convinced of the
feasibility of cooperative enterprises as such and were seizing the op-
portunity offered by the relief grant to undertake their development
in the Valley. Indeed, as one TVA staff member put it, "the sizeable
expenditures incurred by the Authority do not seem justified by actual
accomplishments in terms of experiments of feasibility; they appear
more understandable if regarded as direct action to improve the gen-
eral welfare in the absence of adequate methods (at the time) for
organizing local initiative to do the job."
22 The Act authorized the President to carry on this work, but this authority was
delegated to TVA by Executive Order 6161, June, 1933.
230 University of California Publications in Culture and Society
Reorientation of the work of the Cooperative Research and Experi-
ment Division necessarily followed, with emphasis placed first on a
study of all cooperatives in the Valley and then on linking actual aid
to cooperatives with the broader programs of the Authority. At the
same time, the division was placed under the wing of the agriculturists,
eventually being renamed Rural Cooperative Research Division. A
series of studies of farmer marketing-purchasing cooperatives, soil
conservation associations, and rural electric cooperatives was under-
taken in conjunction with the land-grant colleges, but by 1943 the
major work of the division centered around the extension of advisory
services to voluntary associations organized in connection with other
TVA projects. The latter included primarily (1) soil conservation asso-
ciations handling Authority fertilizer, which might also develop into
marketing and purchasing cooperatives; (2) land-use associations par-
ticipating in the rental of Authority-owned lands; (3) rural electric
power associations acting as distributors for TVA power ; and (4) asso-
ciations using processes and equipment developed in the course of
research sponsored by the TVA Commerce Department. The division
was small, consisting in 1943 of three specialists in cooperatives and
one accountant. At that time, the yearly budget of the division amounted
to approximately $25,000, and about $250,000 had been expended by
TVA on work with cooperatives since 1934.
This evolution, exhibiting a marked change in direction and emphasis,
is important for the grass-roots approach. Linking the development of
voluntary associations to substantive programs, such as the distribu-
tion of fertilizer and electric power, affords an opportunity for the
participation of local citizens in the administration of those programs.
Random organization of cooperatives as ends in themselves does not
lend itself to that type of administrative partnership or involvement.
COOPTATION IN THE FERTILIZER PROGRAM
The involvement of local citizens in the fertilizer test-demonstration
program 23 takes place largely through the county soil improvement asso-
ciations. 24 The organization of these groups is the responsibility of the
agricultural extension services and is not undertaken by TVA per-
sonnel. The associations select test-demonstration farmers within the
county and distribute TVA phosphates. Freight charges on the latter
must be paid by the local farmers, but this is always accomplished
23 See above, pp. 130-132.
24 Known variously as "soil erosion control," "soil improvement," or "soil conser-
vation" associations.
Selznick : T VA and the Grass Roots 231
through a county association. Carlot shipments are the rule, for TVA
does not consign fertilizer directly to individual farmers. Thus, if only
in order to fulfill this formal requirement, there is a county soil-
improvement association wherever the test-demonstration program oper-
ates 119 of them in the Valley area.
In terms of the grass-roots doctrine, the proliferation of county asso-
ciations through the Valley, at the end points of the fertilizer distribu-
tion process, represents a mechanism whereby every participant in the
program may have a share in making decisions which affect him. How-
ever, in pressing inquiry as to whether the concept of formal cooptation
applies in this situation, it is necessary to look toward the functions
which these organizations serve in fulfilling the needs of those who
organized them. Based upon information provided by participants in
TVA and in the extension service, the following points may be adduced as
suggesting the cooptative character of the soil improvement association :
1. The county associations solved an administrative problem of the
extension service by relieving extension workers of official responsibility
for the handling of money. It is in the interest of the colleges to preserve
the status of county agent personnel as educational rather than admin-
istrative, and in any case the extension service does not have and may
not desire authority to undertake the execution of noneducational func-
tions. The ideal case is that in which the county agent merely advises
local groups of farmers. However, the agent in the field is involved in a
continuous dilemma. He cannot very well accept responsibility for push-
ing a program and at the same time ignore the necessity for imple-
menting it on his own initiative where other hands are lacking; nor
can he accept responsibility and ignore the need for measures of control
over the administrative process even if carried on by others. This dif-
ficulty becomes especially evident when an extension-service worker is
actually paid to carry on a definite program with concomitant pressure
for results, as is true of the assistant county agents who receive their
salaries out of TVA funds. Consequently, there is a strong tendency for
rather tight control over the program to be maintained, utilizing the
county associations as mechanisms for a formal shifting of responsibility.
The role of the extension worker in such circumstances is analogous to
that of the executive secretary of a largely paper organization, who often
is in such complete control and so indispensable to its functioning that
he is the organization.
2. Although often and, especially, early in the program, the extension
agents themselves chose the participants in the test-demonstration work,
232 University of California Publications in Culture and Society
it became apparent that it would be useful to have an organization
through which the selections could be made. Since many farmers thought
of the program as essentially the distribution of "free fertilizer," oppor-
tunities for complaints that the county agent's office was "playing fa-
vorites" would certainly arise. Thus if choices were formally made
through some sort of farmer committee, complainants could be referred
to the committee and thus responsibility would be shunted away from
the county agent. At the same time, through his personal influence and
as a result of a general apathy, the agent would be able to see to it that
the farmers he wanted would be included in the program. In many
cases, the action of the association on selections represented simply a
formal confirmation of the choices of the agent. This does not mean that
the county agent always desired to make his own choices. On the con-
trary, the association device would be welcome to him precisely because
it might operate to relieve him from the pressure of influential individ-
uals. However, though he might prefer to leave selections solely to the
farmers as a group, responsibility for results would necessarily force
him to intervene in order to see that a sufficient number of reliable
elements were included to assure some success to the program in the
county. Here we have an example of the principle (see below, p. 261)
that formal cooptation requires the exercise of informal control over
the coopted elements.
3. Indicative of the character of the county associations as adminis-
trative arms of the extension service is that the organization of such
associations is provided for in project agreements between TVA and
the agricultural colleges, but the soil improvement association itself is
never a party to such an agreement. The project agreements lay down
policy governing the test-demonstration program within the state, and
it would seem that if the local groups of citizens are actually to partici-
pate in the determination of policy it would be necessary for that parti-
cipation to reach above the county level. But it is not readily conceivable
that the extension service would accept responsibility for the execution
of the test-demonstration program and at the same time agree to sig-
nificant participation by a major third party. In general, references to
the county associations in administrative documents support the idea
that they are thought of as tools utilized by a particular agency for the
implementation of its program.
4. There is a general and readily observable identification of the
machinery of the county agent system with the local associations. The
headquarters of the local association is normally the extension service
Selznick : T VA and the Grass Roots 233
office in the county courthouse, its records are in the county agent's files,
its gatherings are under his auspices. In 1942, after eight years of opera-
tion of the test-demonstration program, report forms filled out in con-
nection with a survey of the progress of the soil improvement associations
were in most cases signed by the assistant county agents. The official
Test-Demonstration Information Book, published in 1943, informs us
that "early capital for many associations was acquired from a handling
charge of a few cents on each bag of phosphate issued to demonstrators.
Such funds are widely used for educational purposes such as the pur-
chase of cameras, projectors, film, and slides for the county agent's
office." (Emphasis supplied.)
5. Perhaps the crucial test of control over coop ted citizens' groups is
this : is access to the association by outside elements channeled through
officials of the coop ting agency? If it is understood that accessibility is a
primary objective of the cooptative mechanism, then it may be antici-
pated that such organizations will be looked upon as the "property" of
their initiators. By common consent, rights in this property will be
respected by officials of other agencies who in their turn will expect
a similar courtesy. 25 In general, it is commonly admitted that the soil
improvement associations of the TVA-extension service program belong
to the county agent's office and will not be approached save through the
good offices of the extension worker in charge. This is not a result of
undemocratic attitudes but is simply because the associations were origi-
nally organized to serve certain administrative functions. An investment
of time and energy was made, with the expectation that certain program
responsibilities would be fulfilled. It would be idle to expect that the
official in charge would be willing for an organization set up in response
to his needs to be removed from his influence or diverted toward projects
inimical to his purposes.
These considerations underline the cooptative character of the volun-
25 In December, 1942, the Forestry Eelations Department of TVA explored the
possibility of using county program planning committees in the development of
Civilian Conservation Corps camp work loads. In reviewing the incident, a forestry
official commented : "Since these committees are organizations sponsored and directed
by the state extension services, it was deemed advisable to secure the consent of the
state extension service in dealing directly with these organizations." The state ex-
tension director gave his permission, and stated that he would authorize his district
agents to call meetings of county planning committees at any time such meetings
were desired by the TVA foresters. "He indicated that original contacts should be
made through the district agents, and that thereafter, meetings might be arranged
for through the county agents." We see here an operational criterion for locating
the de facto control of such associations. One function of the voluntary groups is to
control the avenues of organized access to the farmers, thus strengthening the ex-
tension service by making it indispensable to other operating agencies.
234 University of California Publications in Culture and Society
tary groups organized to implement the test-demonstration program.
Though not often formulated explicitly, the facts do not go unrecognized
by the TVA staff. One participant put the matter in these terms, speak-
ing of the early period :
In the early days of the program, there was considerable shyness or aloofness on the
part of farmers as to their willingness to test these experimental fertilizers. The
seven land grant institutions of the Valley states agreed to contact individual farmers
through their farm extension agents in the various counties. Many county agents
when asked to organize the work in their particular counties accepted "all comers"
as unit farm test demonstrators. During the first few years it appeared necessary
for the county agent to run the show. The program was small and it could be efficiently
operated by the county and assistant county agents through personal contacts with
each individual farmer. The county agent and the farmer made the farm plan and
discussed as to how much and where the fertilizer was to be used and other improved
practices. When the total amount of the material exceeded 30 tons or more, the
county agent through the extension service ordered out the shipment. He made
arrangements to pay the freight and when the car arrived he helped to load it on
the test-demonstration farmer's conveyances. If any material was uncalled for he
made the arrangements for its storage. In many instances he in turn collected from
the individual farmer his proportion of the freight, paid the various bills and if a
deficit existed, oft times made up the deficit from his personal funds or if there was
a small surplus, put the surplus away or mingled it with his own funds. These efforts
were truly cooperative ventures one man one vote affairs and the county agent
was the one man one vote.
But as the program expanded, the necessity for functioning through
groups became more pressing :
As the test-demonstration program caught on and neighbors heard about the "free
TVA fertilizer," as it was often called, and about the results which were being ob-
tained, many more farmers wanted to sign up. The introductory period of the program
had passed and it now assumed a different aspect. Test-demonstration farmers began
to convey a definite meaning, certain procedures were developing. Committees were
set up for the purpose of administering the program. These committees were not
particularly active at first, but at least they went through the motions of selecting
new demonstrators and approving requisitions for phosphate, primarily upon the
recommendation of the farm agent. Even at this stage the county agent still actually
did the selecting but the responsibility for selection had the rubber stamp of approval
of a local committee.
And we may note especially, in terms of the concept of cooptation :
In many ways the committee was convenient and useful as well as being a f agade.
Their names added prestige to the program. They often could handle complaints more
tactfully and with better results than could the agent. They helped to publicize the
activity and their farms often already above the average farms would be desirable
to visit on their tours. Quite often the records of the committeemen were better kept ;
they had a better understanding of what was being attempted and as better farmers
Selznick : T VA and the Grass Roots 235
were getting better results. Nothing succeeds like success and to the results obtained
by many of these first test-demonstration committeemen the county agent could point
with considerable pride.
Although these remarks pertain especially to the early years of the pro-
gram, in fact the points adduced above were generally applicable to
most of the associations as late as 1943.
Some officials of TVA headquarters look forward to the development
of a strong cooperative movement based upon the network of soil
improvement associations established under the test-demonstration pro-
gram. Consequently, encouragement has been given to the local associa-
tions to branch out into marketing and purchasing activities in order
that they might be strengthened by these added functions. The widening
of activities would give the local farmers a greater stake in their associa-
tions and provide an impetus for incorporation and for otherwise im-
proving their administration. Some accomplishments along this line have
been noted : handling of lime by associations in Kentucky ; distribution
of the TYA fertilizer by-product, calcium silicate slag, through associa-
tions in northwestern Alabama ; and purchasing and marketing ventures
in northern Georgia through a regional federation embracing several
counties.
On the whole, however, the state of the county associations in 1943
was not very satisfactory to those especially interested in the promotion
of cooperatives. It was felt that the extension service remained hesitant
about organizing unlimited membership organizations for fear that such
organizations would "get out of their hands." 28 Of the 119 county associa-
tions in the Valley in 1943, a large proportion were thought to be more
or less inactive, and their total net worth, less than $200,000, to be rela-
tively small.
In addition to the county soil improvement associations, voluntary
groupings at the community level have been organized. The watershed
area projects, 27 another phase of the test-demonstration program, involve
the distribution of fertilizer to entire farming communities. The forma-
tion of some community group, and the holding of at least a few initial
meetings, is a condition of participation in the program. One of the tasks
assigned to the organization is that of conducting a social and economic
community self -survey, which theoretically is useful in setting a base
line from which progress may be measured. Like the county associations,
the community groups are formed at the initiative of the extension-
26 See above, p. 145, relative to the Farm Bureau's attitude toward cooperatives.
27 See above, p. 132.
236 University of California Publications in Culture and Society
service workers and normally require considerable effort to be main-
tained.
Agricultural extension workers in the field consider the community
organizations as effective means of furthering their work. Precisely be-
cause of their utility, and because of the large investment of time and
energy required to activate them, there is a strong tendency to retain a
considerable influence over the local groups. In fact, most of the groups
are so weak, and require so much attention to be kept alive, that influ-
ence by the county agent's office is taken for granted. Investigation car-
ried on in one Tennessee county one of the better counties in the
Valley revealed that the assistant county agents assigned to the test-
demonstration and readjustment programs were the mainstay of the
community groups. One of these workers found it necessary to build up
his community meetings by visiting key people several hours before the
meeting in order to insure at least a minimal attendance. Conversations
with local community leaders indicated that they did not think of the
meetings as theirs, and came primarily for information, for phosphate
clearance, and to some extent to please the extension agent. With some
exceptions, there was observed a tendency for the associations to be
primarily farmers' clubs, lacking links to other community groups. 28
An assistant county agent in the county just referred to pointed
out that the community meetings were sometimes mere formalities, re-
quired for the approval of farmers' requests to participate in the test-
demonstration program. Such formal approval was required by TVA,
and without such a requirement it is possible that fewer community
meetings would have been held. Another agent in the same county said
emphatically that he would not want the community organizations to
become independent of his influence. While he viewed his role as pri-
marily educational, nevertheless he felt that the continuation of a good
program depended upon continued intervention and a measure of con-
trol on his part.
There is one generalization which emerges from a study of the TVA-
extension service community organization work which helps to specify
the conditions under which participation by local people is meaningful
or is instead reduced to mere involvement. In many areas of the Ten-
28 Extension workers were often sincerely distressed by the weakness of the com-
munity organizations, and certainly were doing their best to strengthen them ; but
it appears that the pressure of responsibility for the administration of a definite
program rather than for community organization work as such was often decisive.
Lack of awareness of the elements of rural social organization may also have been a
factor, in addition to the barrenness of initial social organization within some of
the communities.
SelznicJc : T VA and the Grass Roots 237
nessee Valley traditional patterns of cooperation at the community level
are only poorly developed, so that when new organizations are initiated
they tend to remain external to the life of the community. When, how-
ever, the community possesses a history of cooperative effort, and a long
acquaintance with its machinery, a new formal organization may be
caught up within the older network and provided with a leadership and
with well-established channels of access to the people of the area. This
was notably true of a community in a mountain county of Tennessee
which for many years had benefited from community organization work
carried on by a Presbyterian home missionary and later by a pastor of
some reputation. After a generation of such activity, the community
was in a position to make the TVA fertilizer testing program an integral
part of its established cooperative efforts. This meant that the commu-
nity itself exercised a far stronger role than most vis a vis the extension
worker. It may therefore be suggested as a hypothesis that the strength
in terms of local control and meaningful participation of voluntary
formal organizations established to implement administrative objectives
will depend upon the initial social organization of the community. 29
In the fertilizer distribution program, the voluntary associations as
such do not appear to have operated significantly to build up loyalty
to the TVA as an organization. The phosphate program has certainly
been welcomed, and has created good feeling, but TVA has done as much
as possible to play down its own role in favor of the extension service.
Most farmers, however, are either aware of the TVA's role despite its
lack of advertising or fail to distinguish the various brands of "govern-
ment men." From the point of view of grass-roots attitudes and public
support it seems fair to suggest that the readjustment program, which
concentrates personnel and phosphates in areas most seriously affected
by the impoundment of the reservoirs, does have a noticeable effect upon
attitudes toward the Authority. Antagonism toward TVA at the grass
roots often came from farmers whose lands were flooded and who had
to move. But careful attention to the needs of those farmers over a period
of years, and especially the concentration of advisory services and mate-
rial aids in their direction, has mollified some of those early antagonisms.
^ Of course, initial social organization may also be a hindrance. If a new program
is identified with existing organizations or leaders within the community, uninte-
grated elements may remain such and the community in its entirely will not be
reached. Social structure offers a ready highway for effective administration, but
its rigidities will usually require that some toll be paid. Moreover, to the extent that
local control increases, the program of the administrator may be placed in jeopardy,
for there is no guarantee that the vehicle he has used will not be taken out of his
hands and all his work subverted.
238 University of California Publications in Culture and Society
However, it is not the associations as such which have accomplished that,
though they probably do add a local flavor to the programs and provide
a means for the involvement of leading individuals in their adminis-
tration.
KURAL ELECTRIFICATION COOPERATIVES
TYA distributes power at wholesale to municipalities, cooperatives, and
large industrial contractors. The Authority's policy is to avoid direct
distribution to the retail consumer wherever possible, preferring to sell
to agencies of local government, such as municipal power boards, or to
voluntary associations. The TVA Act of 1933 directed the agency to
"give preference to States, counties, municipalities, and cooperative
organizations of citizens or farmers, not organized or doing business for
profit, but primarily for the purpose of supplying electricity to its own
citizens or members." 30 In 1943 power was being sold to 83 municipalities
and 45 cooperatives, as well as to such large industrial consumers as
Monsanto Chemical Company and the Aluminum Company of America.
The organization of local associations for the distribution of TVA
power to largely rural areas was begun in 1934. By 1939, contracts were
in effect with 22 such groups, a figure which was more than doubled
three years later. It is not contended by TVA that the cooperatives origi-
nated in a completely spontaneous movement, although there were al-
ways some local leaders who took the initiative. Speed was an important
factor during the early period, not only because of pressure to get the
program into operation to show some results, but because of the exigen-
cies of controversy with the private power companies. "Spite lines" were
being built, 31 and it was necessary to launch some of the associations very
quickly in order to outmaneuver the power companies. This left little
opportunity for full participation by the prospective membership, and
little time for education.
Although the electrical cooperatives are independent organizations,
relations to the Authority are very close. The contracts for sale of power
30 48 Stat. 58 (1933) See. 10. Electrical cooperatives established under the TVA
program are not essentially different from similar cooperatives set up in other parts
of the country as a result of the work of Rural Electrification Administration. In-
deed, the EEA has also operated in the Valley, typically with a cooperative to which
TVA functions as power wholesaler and EEA as financing agency.
81 The construction of lines into rural areas by private companies approximately
paralleling lines proposed by a cooperative, thus undermining the possibility of
successful financial operation by the cooperative. See "Eural Electrification Adminis-
tration Activities in the Tennessee Valley Area," statement by W. E. Herring,
Consulting Engineer, Eural Electrification Administration, Reports and Exhibits of
the Joint Committee Investigating the Tennessee Valley Authority, Senate Document
No. 56, Pt. 2, 76th Cong., 1st sess. (1939).
Selznick: T VA and the Grass Roots 239
at wholesale are stringent, and include specification of rates at which
the cooperative may resell the electricity at retail as well as stipulations
with regard to finances and organization. Of course, TVA's punitive
powers in case of infringements of the contracts are not very great in
practice ; the TVA would certainly hesitate to shut off electricity sup-
plied to farmers simply because of the recalcitrance of officials of a
cooperative. However, in the case of the cooperatives, the nature of the
administrative relations makes for a large degree of influence on the part
of TVA. 32 Many cooperatives are not strong enough to stand alone, and
are dependent upon TVA for aid, as in setting up bookkeeping systems.
Managers of the cooperatives are often selected jointly by TVA and RE A.
The Authority inevitably accepts some public responsibility for the
cooperatives, in terms of both financial success and public relations. It
is therefore concerned that the cooperatives should take adequate meas-
ures to achieve success in both fields. Hence Authority personnel have
initiated studies which might aid the cooperatives in maintaining their
tax-exempt status; and officials of the TVA Personnel Department
have attempted to apply some pressure to bring the cooperatives into
line with TVA labor relations policy, especially as that relates to labor-
management cooperation and the recognition of the International
Brotherhood of Electrical Workers. But in facing these problems the
Authority is involved in a dilemma : if it is actually to influence person-
nel policy and other phases of the organization of the cooperatives, it may
be necessary to set up machinery for doing so; at the same time, to
establish such machinery would strengthen Authority control and thus
vitiate the original purposes for which the cooperative device was
promoted. Circumstances have not demanded a choice between these
alternatives, however, and the Authority has contented itself with in-
formal controls beyond the terms of the contracts.
TVA officials carry on some educational work among the cooperatives,
oriented in large part toward broadening the understanding of both
officials and members in the principles of cooperative organization. 33
32 This is apparently not true of the relations between TVA and the municipalities.
Indeed, one observer reported in 1939 that there existed among TVA staff members
"a decided dislike for the municipalities and a decided preference for the coopera-
tives." This was attributed to the feeling that "it is easier to deal with the co-ops
because of the absence of politics, a greater willingness to accept TVA suggestions,
and no desire on the part of the co-ops to make money for a general fund to reduce
taxes." The relatively greater influence exercised by TVA over the cooperatives than
over the municipalities was stressed by the manager of one of the large municipal
power boards.
33 Some of the educational work is simple trouble shooting, such as may be called
for when there is a local movement in a small community for the establishment of
240 University of California Publications in Culture and Society
There is said to be some difference in outlook within the Authority, how-
ever. Representatives of the Power Utilization Department reputedly
interpret membership relations as public relations and are more inter-
ested in increasing electricity sales than in the development of Kochdale
cooperatives.
Reverting again to the problem of approach to the voluntary associa-
tion as an operational test of control, it appears that there is a tendency
for representatives of other agencies to contact the electric cooperatives
through TVA. However, while TVA would certainly be concerned about
contacts inimical to its own good relations with the cooperatives, it
does not exert the same close control over them in this situation as the
county extension service agents do over the county soil improvement
associations.
In general, people who buy electricity from local cooperatives in the
Valley think of these organizations as "TVA co-ops." This is probably
due in part to the wide publicity which attended the introduction of
TVA and more particularly because, in some areas now serviced by
cooperatives, TVA did sell directly to the consumer during the early
period. Identification of the cooperatives with TVA is made easier by
slight participation of the local people in the activities of the coopera-
tives. The members generally function as consumers of electricity and
little more. Attendance at annual meetings is not large, and usually
some inducement refreshments, or perhaps a prize contest is neces-
sary. Most of the cooperatives require a five per cent attendance at an-
nual meetings to constitute a quorum, but in 1942 one-fourth of the
associations did not meet this requirement. Moreover, from the point
of view of meaningful participation, annual meetings may not amount to
much more than rallies. The monopoly character of electric power dis-
tribution and the dearth of personal contact between members and
officials contribute to lack of participation. 34
As might be expected, there is some tendency for the development of
manager-director oligarchies within the cooperatives. Some managers
an independent distributing system. TVA might consider such a proposal, if formu-
lated, unfeasible ; yet it would be embarrassing to have to refuse to sell electricity
to a group which wanted it. In this connection, one may note that the Authority
found itself in disagreement with EEA as to the feasibility of certain proposed
cooperatives in the Valley but was hardly in a position to deny electric service to
several thousand farmers on that account.
34 It seems doubtful that, during normal periods, the routine role of consumer
represents an adequate basis for organizational participation. It may be questioned
further whether it is always desirable to summon the participatory energies of a
community without regard to the problems which present themselves to the people
as real issues, or even to the limitations of such energies.
Selznick: TVA and the Grass Roots 241
think of the members as customers, and many are reportedly director-
controlled, with a tendency toward no contest at elections. The number
of nominees at annual meetings will sometimes equal the number of
directors, indicating that some prior decision has been made. There is
also a tendency for some of the cooperatives to be run by their attorneys.
Since many of the questions which come up are legal, the manager may
call the attorney rather than the directors for advice and consultation.
Thus a close relation between manager and attorney may grow up, with
the latter exerting a strong informal influence.
A study of fifteen of the electrical cooperatives, concluded in 1939,
made the following point :
From a strictly cooperative point of view, the outstanding weakness of the associa-
tions studied is the lack of educational work among the membership. No association
has an adequate educational program; all have dealt slightingly with cooperative
membership education and membership relations. Association officers have not taken
advantage of the opportunity to build up membership interest, loyalty, and a con-
sciousness of ownership and pride among their members. If there were a fuller under-
standing among the membership of rural electrification through cooperatives, the
load-building program would be easier and more successful. Unless more is done in
the future than in the past on membership education, serious difficulties may arise
which are likely to affect the actual existence of these associations as cooperative elec-
tricity-distributing units.
Here again we have a situation in which the pressure to get things done
serves to reduce the organization from one having democratic implica-
tions to one which functions as a device for the accomplishment of
administrative tasks.
It should be noted that precisely because the electrical cooperatives
do have a definite leadership of their own ; because they do have their
own offices, hire their own employees, keep their own records ; because
they do make significant decisions f or example, in regard to the exten-
sion of their lines ; because they have relations with RE A as well as TVA ;
for these reasons it is impossible to think of them as so weak and de-
pendent as are the county soil improvement associations described above.
In general, the electrical cooperatives, though not completely independ-
ent, are considered to be fairly strong. The net worth of the electrical
cooperatives in 1943 was estimated at $3,000,000.
One comment which has been made on TVA's orientation toward co-
operatives refers to what appears to be an exclusively rural perspective.
The main group of cooperative specialists is bracketed within the De-
partment of Agricultural Relations. The latter, being heavily committed
to channeling all programs through the agricultural extension service,
242 University of California Publications in Culture and Society
exercises an inevitable inhibition on the field within which the coopera-
tive specialists may work. The problem of ultimately overcoming the
rural-urban distinction in administrative approach would seem to be
especially pressing in relation to the electrical cooperatives. One of the
basic over-all problems in the TVA power program has been to achieve
a proper combination of territory in the distributing system so that
urban areas might help support the cost of rural lines. The Authority,
in deciding against a unified distributing system, compromised this issue
at the outset, since presumably a unified system would have permitted
big-city profits to be plowed into rural lines. Doubtless the political cir-
cumstances which shaped the inception of the program, and perhaps the
intent of the Act, made that decision inescapable. But it seems apparent
that administration of the program at the grass roots, through independ-
ent local agencies, has made it especially difficult for the Authority to
take positive action on the question of persuading municipalities to
extend their lines into rural areas.
OTHER VOLUNTARY ASSOCIATIONS
Two additional situations in which the Authority has sponsored volun-
tary associations as means of involving local people in the administration
of its programs will be reviewed briefly.
Associations Handling Rental of TVA Lands. The Authority con-
trols large tracts of land adjacent to its reservoirs, amounting to almost
400,000 acres in 1943. Those portions of the land holdings considered
suitable for agricultural use are leased to local farmers as part of the
readjustment program. The program is channeled through the exten-
sion service; here again, as in fertilizer distribution, local associations
make assignments of land and handle the money involved. In some areas,
the same county soil conservation associations which handle the test-
demonstration activities are utilized, but in others a new device has been
introduced a multiple-county association operating for an entire reser-
voir area: the Kentucky Reservoir Land-Use Association, the Norris
Reservoir Area Land-Use Association, and the Chickamauga Land-Use
Association. The associations classify the land thus indirectly deter-
mining rental values and receive and act on bids for licensing. Ten per
cent of the licensing fees are retained by the associations to cover ex-
penses.
The local organizations are not entirely independent, however. They
receive guidance from extension service representatives, most of whom
work out of the county agent's office with their salaries paid by TVA
Selznick : T VA and the Grass Roots 243
under the readjustment program. 35 In effect, the associations are crea-
tures of the extension service. They are not actually cooperatives, al-
though that term appears in their charters, since the farmers who
participate in the rental operations are not members. The Norris and
Chickamauga associations' membership consists of the county agricul-
tural planning committees of the constituent counties. But since the
planning committees are actually sponsored and directed by the exten-
sion service, 38 the relation of the multiple-county organizations to the
extension service remains the same. Contact with the associations is al-
ways through the extension service, a rule which holds though not
uniformly for TVA officials as well as for outside agencies.
Audits conducted by the TVA Finance Department in 1940-1941
revealed certain discrepancies in the activities of the associations, such
as leasing to local businessmen financially able to pay large rentals in
advance and to employ farm labor to operate the land. To those making
the audit, this seemed detrimental to the social objectives of the Author-
ity's agricultural program. However, it is significant that in one Ala-
bama county in which such leases were made, those with grass-roots
perspectives justified the practice on the theory that leasing to absentee
farmers did not alter the conditions which prevailed before the Author-
ity bought the land. Since the local extension service personnel concurred
in this judgment, we see here another example of the dilemma faced by
TVA in turning over the administration of its program to grass-roots
agencies. It is not known to what extent the eligibility of lessees for bene-
fit payments under the Agricultural Adjustment Administration pro-
gram has affected the licensing procedure. Sometimes AAA payments
may have exceeded the cost of renting the land; it seems likely that
under such circumstances there would be pressures which might result
in depriving needy farmers of the use of the land.
In at least one of the reservoir areas, it was noted that there was
considerable participation and independence on the part of community
and county committeemen engaged in the leasing operations. Commit-
teemen are paid by the day for their services, and the need for the land
by the local population affords them opportunities for improving farm
practices and for stimulating interest in rural planning. 87
35 There were 35 assistant county agents in "relocation, adjustment and land
licensing" in 1943. These men also work on the test-demonstration program, which
is intensified in the reservoir-affected counties.
88 See above, p. 233, n. 25.
87 The legal requirement that licensing be limited to periods of one year is noted
by field workers as an obstacle to long-time planning, and this is probably one
reason for the informal circumvention of competitive bidding for licenses.
244 University of California Publications in Culture and Society
Town Planning Commissions. In 1938, a program of planning as-
sistance to local communities was initiated by the Authority in coopera-
tion with the State Planning Commissions of Alabama and Tennessee.
This program is administered by an Urban Community Kelations Divi-
sion in the TVA's Regional Studies Department. 38 Two broad purposes
are envisioned: (1) Execution of the Authority's responsibility for
aiding in the readjustment of reservoir-affected communities. Impound-
ing of large bodies of water behind the dams necessitates community
readjustments f or example, to the creation of a new lake front which
should be properly integrated into a plan for the city or town as a whole.
(2) The development of a project to demonstrate types of community
action which will promote the general welfare: more specifically, to
secure the establishment of planning as a recognized function of local
government in the Valley.
Entering a situation in which the planning functions of local govern-
ment were only poorly developed, the Authority has attempted to pro-
vide leadership in this direction by stimulating and aiding local planning
programs of state planning commissions until public interest is devel-
oped sufficiently to insure the maintenance and expansion of the program
without TVA sponsorship. TVA personnel also render direct aid by
providing technical planning services to small communities, assistance
in preparation of model planning legislation, and instructional mate-
rials in planning for school curricula.
The TVA has applied its grass-roots concept in this field by working
through agencies of the state governments and by supporting the crea-
tion of voluntary citizens' groups to serve as official planning commis-
sions for local communities. However, it remains to the state agencies to
guide the local commissions. The state commissions only Alabama and
Tennessee during the period up to 1943 may employ resident plan-
ning technicians whose salaries are reimbursed by TVA, or sometimes
TVA will assign one of its own staff members to work under the com-
mission. Activities are directed primarily toward the small communities
which do not have resources to employ their own technicians and there-
fore must rely upon the state or TVA for this service. In 1943, twenty-
88 It was estimated that by 1943 the TVA had spent about $100,000 on its planning
assistance program, which had stimulated the states to spend $25,000 and the local
communities another $25,000. These are approximations, and represent five years of
operation. The TVA figure does not include the salary of the chief of the Urban
Community Eelations Division, who has other responsibilities ; it must also be noted
that expenditures by the Authority are not strictly comparable to those by the states
because some of the money represents time spent by TVA staff members on prob-
lems of interest peculiar to TVA.
Selznick : T VA and the Grass Roots 245
nine communities and planning commissions in Alabama and Tennessee
were cooperating in the program. In reservoir-affected communities, the
local commissions received guidance in waterfront development and
readjustment, development of commercial, industrial, river terminal,
and recreation facilities, and in comprehensive community planning and
zoning ; in war-affected communities, interest centered on programming
and site selection of war housing, studies of school and recreation needs,
and planning to relate emergency developments to long-time community
needs.
The role of TVA in this activity is perhaps best defined by recalling
that it has felt the need to stimulate local interest in planning. Conse-
quently, though the program may be required by objective community
needs, it can hardly be said to have been a response to needs felt at the
grass roots. Authority personnel helped to draft enabling legislation for
the states and organize the state planning machinery. 39 The relation be-
tween TVA officials and the technicians formally employed by the state
commissions is very close. Indeed, there is some tendency for the latter
to form a de facto field force of the TVA. If TVA has some particular
need, it can operate through these men ; thus, for practical purposes,
there was reportedly little difference in the status of two planning tech-
nicians working in Tennessee, one of whom was a TVA employee, the
other formally employed by the state commission under contract with
TVA. In this program, the TVA assumes a definite leadership, which is
inevitably reflected in informal organizational relations. This sometimes
takes the form of a proprietary attitude toward the sponsored groups,
as when the TVA planning officials refused permission to an independent
investigator to interview members of local planning commissions, be-
cause of doubts and skepticism which might be raised in their minds
by his questions. The need to go through channels TVA or the state
commissions or perhaps both in contacting local groups is, in this case
as in others we have mentioned, an indication of the cooptative char-
acter of the local groups. Having been set up for the accomplishment of
certain program responsibilities, it is considered undesirable to place
those objectives in jeopardy by permitting freedom of access by other
individuals and agencies. 40
39 The Tennessee State Planning Commission is said to have been set up in the office
of TVA's Eegional Planning Director. It was formed from a committee established
by the Tennessee legislature in 1933 to assist the work of the TVA.
40 Another instance, this time involving a state commission rather than TVA, oc-
curred when a representative of a private organization was rebuffed in his attempt
to contact officials of a local community through the director of one of the state
commissions. The director did not like the manner in which his assistance was sought
246 University of California Publications in Culture and Society
The organizational role of the local groups is emphasized further by
the conscious choice made between two possible approaches : the plan-
ning personnel to be sponsored might be either a single planning engineer
attached to an administrative office of the local government or it might
be a planning commission. One of the reasons for preferring the latter
was the belief that the commission would help to win public support for
the planning program. In small communities especially, a nucleus of
informed citizens might have considerable influence in this direction.
Indeed, the public relations problem has supplied an underlying admin-
istrative rationale for the entire grass-roots approach in this field. Early
in the program, it was found that TVA's attempt to work directly in
community readjustment was not well received. The local citizens or
presumably some vocal portion of them appear to have been suspicious
of TVA because the same agency which was responsible for the damage
was giving advice on readjustment to it. The device of working through
state and local commissions, which could take responsibility for plan-
ning, was a convenient answer to this problem.* 1
and so refused to cooperate. Interesting here is that it appears to have been under-
stood on all sides that the local groups should not be approached directly, but rather
through the sponsoring agency.
41 It is not contended that this is the only or the most important reason for the
use of the grass-roots approach. Probably more significant is the professional interest
on the part of planners to build up local understanding of and machinery for town
planning. However, we are concerned to make explicit the underlying organizational
needs which are served by the grass-roots policy.
Conclusion
GUIDING PRINCIPLES AND INTERPRETATION
A SUMMARY
CONCLUSION
GUIDING PRINCIPLES AND INTERPRETATION
A SUMMARY
The entire science considered as a body of formulae having coherent relations to one
another is just a system of possible predicates that is, of possible standpoints or
methods to be employed in qualifying some particular experience whose nature
or meaning is unclear to us. 1 JoHN DEWEY
IT is BELIEVED that the interpretation set forth in the preceding chapters
provides a substantially correct picture of a significant aspect of the
TVA's grass-roots policy at work. Far from remote, or divorced from
what is considered pertinent by informed participants, the analysis
reflects what is obvious to those who "know the score" in TVA. 2 Of course,
this exposition is more explicit and systematic, and the relevant impli-
cations are more fully drawn out, but in main outline it can come as no
surprise to leading officials of the Authority. This is not to suggest that
there are no errors of detail, perhaps even of important detail. The
nature of this kind of research precludes any full assurance on that.
While much of the material is derived from documentary (though
largely unpublished) sources, much is also based upon interviews with
members of the organization and with those nonmembers who were in a
position to be informed. Care was taken to rely upon only those who had
an intimate, as opposed to hearsay, acquaintance with the events and
personalities involved. Those who are familiar with the shadowland of
maneuver in large organizations will appreciate the difficulties, and the
extent to which ultimate reliability depends upon the ability of the in-
vestigator to make the necessary discriminations. They will also recog-
nize the need for insight and imagination if the significance of behavior,
as it responds to structural constraints, is to be grasped. All this involves
considerable risk.
If the use of personal interviews, gossip channels, working papers, 3
x John Dewey, Problems of Men (New York: Philosophical Library, 1946),
p. 221.
2 Although responsibility for the analysis rests solely with the author, it should
be emphasized that this study was made possible by the willingness of TVA to make
its records and personnel available. This is a happy precedent which we may hope will
be followed by other organizations, public and private.
3 Some of the materials quoted in the study are unofficial in the sense that they
would be vigorously edited before receiving even the public status of a memorandum
sent to another department within TVA. This would be so with comparable docu-
ments in any large organization, public or private.
[249]
250 University of California Publications in Culture and Society
and participation* opens the way for error, it remains, however, the only
way in which this type of sociological research can be carried on. A care-
ful investigator can minimize error by such means as checking verbal
statements against the documentary record, appraising the consistency
of information supplied to him, and avoiding reliance on any single
source. On the other hand, he will not restrict his data to that which is
publicly acknowledged.
The possibilities of factual error, however great, are probably less
important as hazard than the theoretical orientation of the study. To
be sure, an empirical analysis of a particular organization, of its doc-
trine, of a phase of policy in action, of its interaction with other struc-
tures, was our objective. But in order to trace the dynamics of these
events, it has been necessary to attempt a reconstruction, which is to say,
a theory, of the conditions and forces which appear to have shaped the
behavior of key participants.
Theoretical inquiry, when it is centered upon a particular historical
structure or event, is always hazardous. This is due to the continuous
tension between concern for a full grasp and interpretation of the mate-
rials under investigation as history, and special concern for the induction
of abstract and general relations. Abstractions deal harshly with "the
facts," choosing such emphases and highlighting such characteristics as
may seem factitious, or at least distorted, to those who have a stake in
an historically well-rounded apprehension of the events themselves. This
is especially true in the analysis of individual personalities or social
institutions, for these demand to be treated as wholes, with reference
to their own central motives and purposes, rather than as occasions for
the development of theoretical systems. This general, and perhaps ines-
capable, source of misunderstanding being admitted, let us review the
concepts which have been used to order the materials of our inquiry.
SOCIOLOGICAL DIRECTIVES
This volume has been subtitled "A Study in the Sociology of Formal
Organization." This means that the inquiry which it reports was shaped
by sociological directives, more especially by a frame of reference for
the theory of organization. 6 These directives are operationally relevant
without, however, functioning as surrogates for inductive theory itself.
* The author spent most of his year's stay at TVA in daily contact with personnel
of the agency. A number of weeks was spent in intensive contact with extension
service personnel in the field.
5 For a fuller statement than the summary which follows see Philip Selznick,
"Foundations of the Theory of Organization," American Sociological Review, XIII
(February, 1948).
Selznick : T VA and the Grass Roots 251
That is, while they provide criteria of significance, they do not tell us
what is significant; while they provide tools for discrimination, they
do not demand any special conclusions about the materials under inves-
tigation. 6 The fundamental elements of this frame of reference are these :
1. All formal organizations are molded by forces tangential to their
rationally ordered structures and stated goals. Every formal organiza-
tion trade union, political party, army, corporation, etc. attempts
to mobilize human and technical resources as means for the achieve-
ment of its ends. However, the individuals within the system tend to
resist being treated as means. They interact as wholes, bringing to bear
their own special problems and purposes ; moreover, the organization
is imbedded in an institutional matrix and is therefore subject to pres-
sures upon it from its environment, to which some general adjustment
must be made. As a result, the organization may be significantly viewed
as an adaptive social structure, facing problems which arise simply be-
cause it exists as an organization in an institutional environment, inde-
pendently of the special (economic, military, political) goals which
called it into being.
2. It follows that there will develop an informal structure within the
organization which will reflect the spontaneous efforts of individuals
and subgroups to control the conditions of their existence. There will
also develop informal lines of communication and control to and from
other organizations within the environment. It is to these informal rela-
tions and structures that the attention of the sociologist will be primarily
directed. He will look upon the formal structure, e.g., the official chain
of command, as the special environment within and in relation to which
the informal structure is built. He will search out the evolution of formal
relations out of the informal ones. 7
3. The informal structure will be at once indispensable to and conse-
quential for the formal system of delegation and control itself. "Wherever
command over the responses of individuals is desired, some approach
in terms of the spontaneous organization of loyalty and interest will be
8 Thus, while approaching his materials within a guiding frame of reference,
the author was not committed by this framework to any special hypothesis about the
actual events. Indeed, he began his work with the hypothesis that informally the
grass-roots policy would mean domination by TVA, because of its resources, energy,
and program. After the first two months in the field, however, this hypothesis was
abandoned as a major illuminating notion.
7 For discussion of informal organization, see F. J. Roethlisberger and W. J. Dick-
son, Management and the Worker (Cambridge: Harvard University Press, 1941),
pp. 524 ff; also Chester I. Barnard, The Functions of the Executive (Cambridge:
Harvard University Press, 1938), chap, ix; Wilbert E. Moore, Industrial Relations
and the Social Order (New York: The Macmillan Co., 1946), chap. xv.
252 University of California Publications in Culture and Society
necessary. In practice this means that the informal structure will be
useful to the leadership and effective as a means of communication and
persuasion. At the same time, it can be anticipated that some price will
be paid in the shape of a distribution of power or adjustment of policy.
4. Adaptive social structures are to be analyzed in structural-
functional terms. 8 This means that contemporary and variable behavior
is related to a presumptively stable system of needs 9 and mechanisms.
Every such structure has a set of basic needs and develops systematic
means of self-defense. Observable organizational behavior is deemed
explained within this frame of reference when it may be interpreted
(and the interpretation confirmed) as a response to specified needs.
Where significant, the adaptation is dynamic in the sense that the utiliza-
tion of self -defensive mechanisms results in structural transformations
of the organization itself. The needs in question are organizational, not
individual, and include : the security of the organization as a whole in
relation to social forces in its environment ; the stability of the lines of
authority and communication ; the stability of informal relations within
the organization ; the continuity of policy and of the sources of its deter-
mination ; a homogeneity of outlook with respect to the meaning and
role of the organization.
5. Analysis is directed to the internal relevance of organizational be-
havior. The execution of policy is viewed in terms of its effect upon the
organization itself and its relations with others. This will tend to make
the analysis inadequate as a report of program achievement, since that
will be deemphasized in the interests of the purely organizational con-
sequences of choice among alternatives in discretionary action.
6. Attention being focused on the structural conditions which influ-
ence behavior, we are directed to emphasize constraints, the limitation
of alternatives imposed by the system upon its participants. This will
tend to give pessimistic overtones to the analysis, since such factors as
good will and intelligence will be deemphasized.
7. As a consequence of the central status of constraint, tensions and
dilemmas will be highlighted. Perhaps the most general source of ten-
8 See Talcott Parsons, "The Present Position and Prospects of Systematic Theory
in Sociology," in George Gurvitch and Wilbert E. Moore (eds.), Twentieth Century
Sociology (New York: Philosophical Library, 1945).
9 As Eobert K. Merton has pointed out to the author, the concept of "basic needs"
in organizational analysis may be open to objections similar to those against the
concept of instinct. To be sure, the needs require independent demonstration; they
should be theoretically grounded independently of imputations from observed re-
sponses. However, we may use the notion of "organizational need" if we understand
that it refers to stable systems of variables which, with respect to many changes in
organizational structure and behavior, are independent.
Selznick : T VA and the Grass Roots 253
sion and paradox in this context may be expressed as the recalcitrance
of the tools of action. Social action is always mediated by human struc-
tures, which generate new centers of need and power and interpose
themselves between the actor and his goal. Commitments to others are
indispensable in action : at the same time, the process of commitment
results in tensions which have always to be overcome.
These principles define a frame of reference, a set of guiding ideas
which at once justify and explain the kind of selection which the sociol-
ogist will make in approaching organizational data. As we review some
of the key concepts utilized in this study, the operational relevance of
this frame of reference will be apparent.
UNANTICIPATED CONSEQUENCES IN
ORGANIZED ACTION
The foregoing review of leading ideas directs our attention to the mean-
ing of events. This leads us away from the problem of origins. 10 For the
meaning of an act may be spelled out in its consequences, and these are
not the same as the factors which called it into being. The meaning of
any given administrative policy will thus require an excursion into the
realm of its effects. These effects ramify widely, and those we select
for study may not always seem relevant to the formal goals in terms of
which the policy was established. Hence the search for meanings may
seem to go rather far afield, from the viewpoint of those concerned only
with the formal program. Any given event, such as the establishment
of a large army cantonment, may have a multitude of effects in different
directions : upon the economy of the area, upon the morals of its inhabit-
ants, upon the pace of life, and so on. The free-lance theorist may seek
out the significance of the event in almost any set of consequences. But
in accordance with the principle stated above, we may distinguish the
random search for meanings which can be, at one extreme, an aesthetic
10 In terms of origins, the TVA's policy though not the grass-roots doctrine qua
doctrine of channeling its agricultural program through the land-grant colleges of
the Valley states may be adequately referred to such factors as the nature of the
formal agricultural program, the resources available for its implementation, and
the administrative rationale which seemed conclusive to leading participants. More-
over, these factors may sustain the continued existence of the policy, and it may
therefore seem superfluous when extraneous factors are brought in and somewhat
tangential explanations are offered. But when we direct our attention to the meaning
of the policy in terms of certain indirect but internally relevant consequences as
for the role of TVA in the agricultural controversy, we have begun to recast our
observation of the policy (taken as a set of events) itself. We are then concerned
not with the question, "how did the grass-roots policy come into being?" but with
the question, "what are the implications of the grass-roots policy for the organiza-
tional position and character of TVA?"
254 University of California Publications in Culture and Society
interest from the inquiry of the organizational analyst. The latter like-
wise selects consequences, but his frame of reference constrains his view :
it is his task to trace such consequences as redound upon the organization
in question ; that is, such effects as have an internal relevance. Thus, only
those consequences of the establishment of the army cantonment in a
given area which result in adjustments of policy or structure in the
administration of the cantonment will be relevant.
There is an obvious and familiar sense in which consequences are
related to action : the articulation of means and ends demands that we
weigh the consequences of alternative courses of action. Here conse-
quences are anticipated. But it is a primary function of sociological
inquiry to uncover systematically the sources of unanticipated conse-
quence in purposive action. 11 This follows from the initial proposition
in our frame of reference : "All formal organizations are molded by
forces tangential to their rationally ordered structures and stated goals"
(p. 251, above). Hence the notion of unanticipated consequence is a key
analytical tool : where unintended effects occur, there is a presumption,
though no assurance, 12 that sociologically identifiable forces are at work.
There are two logically fundamental sources of unanticipated conse-
quence in social action, that is, two conditions which define the inherent
predisposition for unanticipated consequences to occur :
1. The limiting function of the end-in-view. A logically important
but sociologically insignificant source of unanticipated consequence
exists because the aim of action limits the perception of its ramified con-
sequences. 13 This is legitimate and necessary, for not all consequences
are relevant to the aim. But here there arises a persistent dilemma. This
very necessity to "keep your eye on the ball" which demands the con-
struction of a rational system explicitly relating means and ends will
11 Consequences unanticipated from the viewpoint of the formal structure are not
necessarily undesired. On the contrary, the result may be a satisfactory adjustment
to internal and external circumstances, upon which the leadership may find it con-
venient to declare that the results were actually intended, though close analysis might
show that this is actually a rationalization. In this type of unintended consequence,
some need is fulfilled. The same unintended consequence may fulfill a need for a part
of the organization and at the same time cause difficulties for the whole, and con-
versely. Many unintended consequences are, of course, sociologically irrelevant. For
an early statement of this general problem, see Eobert K. Merton, "The Unanticipated
Consequences of Purposive Social Action," American Sociological Review, Vol. I
(December, 1936).
12 Where unintended consequences occur due to error, or to individual idiosyncrasy,
they are sociologically irrelevant. However, there is often, though not always, a
systematically nonrational factor at work whose presence is manifested by mistakes
and personality problems.
13 This follows, of course, from the hypothetical, and therefore discriminating and
ordering, status of the end-in-view. See John Dewey, Logic: The Theory of Inquiry
(New York: Henry Holt, 1938), pp. 496-497.
Selznick: T VA and the Grass Roots 255
restrain the actor from taking account of those consequences which in-
directly shape the means and ends of policy. Because of the necessarily
abstract and selective character of the formal criteria of judgment, there
will always be a minimum residue of unanticipated consequence."
2. Commitment as a basic mechanism in the generation of unanti-
cipated consequences. The sociologically significant source of unantici-
pated consequences inherent in the organizational process may be
summed up in the concept of "commitment." This term has been used
throughout this study to focus attention upon the structural conditions
which shape organizational behavior. This is in line with the sociological
directive, stated above, that constraints imposed by the system will be
emphasized. A commitment in social action is an enforced line of action ;
it refers to decision dictated by the force of circumstance with the result
that the free or scientific adjustment of means and ends is effectively
limited. The commitment may be to goals, as where the existence of an
organization in relation to a client public depends on the fulfillment
of certain objectives; 15 or, less obviously, to means, derived from the
recalcitrant nature of the tools at hand. The commitments generated
by the use of self -activating and recalcitrant tools are expressed in the
proliferation of unintended consequences. 19
14 The use of the terms "end-in-view" and "anticipated" may easily lead to the
fallacy of formulating this problem as one of the subjective awareness of the par-
ticipants. This is a serious error. What is really involved is that which is anticipated
or unanticipated by the system of discrimination and judgment which is applied
to the means at hand. This may, and very often does, involve subjective anticipation
or its want, but need not do so. Moreover, the system may be adjusted so as to be
able to take account of factors previously unpredieted and uncontrolled. This addi-
tion of systematically formulated criteria of relevance occurs continuously, as in
the recognition of morale factors in industry. In the situation detailed above, the
high self -consciousness of the American Farm Bureau Federation apparently led it
to anticipate the possible rivalry from a new organization set up under the Agri-
cultural Adjustment Administration, since it took steps to ward off this threat.
See above, p. 161. This is no accidental perspicacity but a result of the systematic
consideration of just such possible consequences from the implementation of new
legislation. However, the tendency to ignore factors not considered by the formal
system not so much subjectively as in regard to the competence of the system to
control them is inherent in the necessities of action and can never be eliminated.
15 As in the T VA's commitment to become a successful electric power business ;
this type of commitment was much milder in the distribution of fertilizer, permitting
adaptation in this field which would contribute to the fulfillment of the prior com-
mitment to electricity.
^ la Our use of the notion of unanticipated consequence assumes that the functional
significance of such consequences is traceable within a specific field of influence and
interaction. Thus price decisions made by a small enterprise affect the market
(cumulatively with others), with ultimate unanticipated and uncontrolled conse-
quence for future pricing decision. This is not an organizational process. When,
however, the retailer builds up good will or makes decisions which will enforce his
dependence upon some manufacturer, these are organizational acts within a theo-
retically controllable field, and are analyzable within the frame of reference set
forth above.
256 University of California Publications in Culture and Society
The types of commitment in organizational behavior identify the
conditions under which a high frequency of unanticipated consequences
may be expected to occur :
i) Commitments enforced by uniquely organizational imperatives.
An organizational system, whatever the need or intent which called it
into being, generates imperatives derived from the need to maintain
the system. We can say that once having taken the organizational road
we are committed to action which will fulfill the requirements of order,
discipline, unity, defense, and consent. These imperatives may demand
measures of adaptation unforeseen by the initiators of the action, and
may, indeed, result in a deflection of their original goals. Thus the tend-
ency to work toward organizational unity will commit the organization
as a whole to a policy originally relevant to only a part of the program.
This becomes especially true where a unifying doctrine is given definite
content by one subgroup : in order to preserve its special interpretation
the subgroup presses for the extension of that interpretation to the
entire organization so that the special content may be institutionalized. 17
ii) Commitments enforced by the social character of the personnel.
The human tools of action come to an organization shaped in special
but systematic ways. Levels of aspiration and training, social ideals,
class interest these and similar factors will have molded the character
of the personnel. This will make staff members resistant to demands
which are inconsistent with their accustomed views and habits; the free-
dom of choice of the employer will be restricted, and he will find it
necessary in some measure to conform to the received views and habits
of the personnel. Thus, in recruiting, failure to take into account initial
commitments induced by special social origins will create a situation
favorable to the generation of unanticipated consequences. The TVA's
agricultural leadership brought with it ideological and organizational
commitments which influenced over-all policy. This was a basically un-
controlled element in the organization. It is noteworthy that where
the character of any organization is self-consciously controlled, recruit-
ment is rigidly qualified by the criterion of social (class, familial, racial)
origin.
iii) Commitments enforced by institutionalization. Because or-
ganizations are, social systems, goals or procedures tend to achieve an
established, value-impregnated status. We say that they become institu-
17 In the TVA, the agriculturists made vigorous efforts to extend their interpre-
tation of the grass-roots policy to the Authority as a whole ; in respect to the federal
government, the TVA attempts to have its special interpretation of administrative
decentralization become general public policy.
Selznick: TVA and the Grass Roots 257
tionalized. Commitment to established patterns is generated, thus again
restricting choice and enforcing special lines of conduct. The attempt
to commit an organization to some course of action utilizes this prin-
ciple when it emphasizes the creation of an established policy, or other
forms of precedent. Further, the tendency of established relations and
procedures to persist and extend themselves, will create the unintended
consequence of committing the organization to greater involvement than
provided for in the initial decision to act. 18 Where policy becomes institu-
tionalized as doctrine, unanalyzed elements will persist, and effective
behavior will be framed in terms of immediate necessities. An official
doctrine whose terms are not operationally relevant will be given con-
tent in action, but this content will be informed by the special interests
and problems of those to whom delegation is made. Hence doctrinal
formulations will tend to reinforce the inherent hazard of delegation. 19
A variation of this situation occurs when the role of participants comes
to overshadow in importance the achievement of formal goals. Action
then becomes irresponsible, with respect to the formal goals, as in the
"fanatical" behavior of the TVA agriculturists. 20
iv) Commitments enforced ~by the social and cultural environment.
Any attempt to intervene in history will, if it is to do more than com-
ment upon events, find it necessary to conform to some general restraints
imposed from without. The organizers of this attempt are committed
to using forms of intervention consistent with the going social structure
and cultural patterns. Those who ascend to power must face a host of
received problems ; shifts in public opinion will demand the reformula-
tion of doctrine; the rise of competing organizations will have to be
faced ; and so on. The institutional context of organizational decision,
when not taken into account, will result in unanticipated consequences.
Thus intervention in a situation charged with conflict will mean that
contending forces will weigh the consequences of that intervention for
their own battle lines. The intervening organization must therefore
qualify decision in terms of an outside controversy into which it is drawn
despite itself. More obviously, the existence of centers of power and
interest in the social environment will set up resistances to, or accept
and shape to some degree, the program of the organization.
v) Commitments enforced ~by the centers of interest generated in the
18 See above, p. 70 f .
19 We have reviewed above, pp. 59-64, the unanalyzed abstractions in TVA's grass-
roots doctrine, which are given content and meaning by the pressure of urgent organ-
izational imperatives.
20 See above, pp. 205 ft.
258 University of California Publications in Culture and Society
course of action. The organizational process continuously generates
subordinate and allied groupings whose leaderships come to have a stake
in the organizational status quo. This* generation of centers of interest is
inherent in the act of delegation. The latter derives its precarious quality
from the necessity to permit discretion in the execution of function or
command. But in the exercise of discretion there is a tendency for deci-
sions to be qualified by the special goals and problems of those to whom
delegation is made. Moreover, in the discretionary behavior of a section
of the apparatus, action is taken in the name of the organization as a
whole ; the latter may then be committed to a policy or course of action
which was not anticipated by its formal program. In other words, the
lack of effective control over the tangential informal goals of individuals
and subgroups within an organization tends to divert it from its initial
path. This holds true whether delegation is to members and parts of a
single organization, or to other organizations, as in the TVA's relation
to the land-grant colleges.
These types of commitment create persistent tensions or dilemmas. 21
In a sense, they set the problems of decision and control, for we have
identified here the key points at which organizational control breaks
down. Operationally, a breakdown of control is evidenced in the genera-
tion of observable unanticipated consequences. This is the same as to say
that significant possibilities inherent in the situation have not been taken
into account. The extension of control, with concomitant minimization
of unintended consequence, is achieved as and if the frame of reference
for theory and action points the way to the significant forces at work.
The problems indicated here are perennial because they reflect the
interplay of more or less irreconcilable commitments : to the goals and
needs of the organization and at the same time to the special demands
of the tools or means at hand. Commitment to the tools of action is indis-
pensable; it is of the nature of these tools to be dynamic and self-
activating ; yet the pursuit of the goals which initiated action demands
continuous effort to control the instruments it has generated. This is a
general source of tension in all action mediated by human, and especially
organizational, tools. v
The systematized commitments of an organization define its character.
Day-to-day decision, relevant to the actual problems met in the transla-
tion of policy into action, create precedents, alliances, effective symbols,
and personal loyalties which transform the organization from a profane,
21 In effect, we have restated here some of the basic points made in the discussion
above of the inherent dilemmas of the TVA doctrine. See pp. 69-74.
Selznick : T VA and the Grass Roots 259
manipulable instrument into something having a sacred status and thus
resistant to treatment simply as a means to some external goal. That is
why organizations are often cast aside when new goals are sought.
The analysis of commitment is thus an effective tool for making ex-
plicit the structural factors relevant to decision in organized action.
Attention is directed to the concrete process of choice, selecting those
factors in the environment of decision which limit alternatives and
enforce uniformities of behavior. When we ask, "To what are we com-
mitted?" we are speaking of the logic of action, not of contractual obli-
gations freely assumed. So long as goals are given, and the impulse to
act persists, there will be a series of enforced lines of action demanded
by the nature of the tools at hand. These commitments may lead to unan-
ticipated consequences resulting in a deflection of original goals. 22
THE COOPTATIVE MECHANISM
The frame of reference stated above includes the directive that organi-
zational behavior be analyzed in terms of organizational response to
organizational need. One such need is specified as "the security of the
organization as a whole in relation to social forces in its environment."
Responses, moreover, are themselves repetitive may be thought of as
mechanisms, following the terminology of analytical psychology in its
analysis of the ego and its mechanisms of defense. One such organiza-
tional mechanism is ideology ; another, which has been the primary focus
of this study, we have termed cooptation. In the Introduction we have
defined this concept as "the process of absorbing new elements into the
leadership or policy-determining structure of an organization as a means
of averting threats to its stability or existence." Further, this general
mechanism assumes two basic forms : Formal cooptation, when there is
a need to establish the legitimacy of authority or the administrative
accessibility of the relevant public ; and informal cooptation, when there
is a need of adjustment to the pressure of specific centers of power within
the community.
Cooptation in administration is a process whereby either power or the
burdens of power, or both, are shared. On the one hand, the actual center
of authority and decision may be shifted or made more inclusive, with
or without any public recognition of the change; on the other hand,
22 The British Labor Party, when it assumed power in 1945, had to accept a large
number of commitments which followed simply from the effort to govern in those
circumstances, independently of its special program. "Meeting a crisis," in a women's
club as well as in a cabinet, is a precondition for the institution of special measures.
To assume leadership is to accept these conditions.
260 University of California Publications in Culture and Society
public responsibility for and participation in the exercise of authority
may be shared with new elements, with or without the actual redistribu-
tion of power itself. The organizational imperatives which define the
need for cooptation arise out of a situation in which formal authority
is actually or potentially in a state of imbalance with respect to its insti-
tutional environment. On the one hand, the formal authority may fail
to reflect the true balance of power within the community ; on the other
hand, it may lack a sense of historical legitimacy, or be unable to mo-
bilize the community for action. Failure to reflect the true balance of
power will necessitate a realistic adjustment to those centers of institu-
tional strength which are in a position to strike organized blows and
thus to enforce concrete demands. This issue may be met by the kind of
cooptation which results in an actual sharing of power. However, the
need for a sense of legitimacy may require an adjustment to the people
in their undifferentiated aspect, in order that a feeling of general ac-
ceptance may be developed. For this purpose, it may not be necessary
actually to share power : the creation of a "front" or the open incorpora-
tion of accepted elements into the structure of the organization may
suffice. In this way, an aura of respectability will be gradually trans-
ferred from the coopted elements to the organization as a whole, and
at the same time a vehicle of administrative accessibility may be estab-
lished.
We may suggest the hypothesis : Cooptation which results in an actual
sharing of power will tend to operate informally, and correlatively,
cooptation oriented toward legitimization or accessibility will tend to
be effected through formal devices. Thus, an opposition party may be
formally coopted into a political administration through such a device
as the appointment of opposition leaders to ministerial posts. This device
may be utilized when an actual sharing of power is envisioned, but it is
especially useful when its object is the creation of public solidarity, the
legitimization of the representativeness of the government. In such cir-
cumstances, the opposition leaders may become the prisoners of the
government, exchanging the hope of future power (through achieving
public credit for holding office in a time of crisis) for the present func-
tion of sharing responsibility for the acts of the administration. The
formal, public character of the cooptation is essential to the end in view.
On the other hand, when cooptation is to fulfill the function of an adjust-
ment to organized centers of institutional power within the community,
it may be necessary to maintain relationships which, however conse-
quential, are informal and covert. If adjustment to specific nucleuses of
Selznick : T VA and the Grass Roots 261
power becomes public, then the legitimacy of the formal authority, as
representative of a theoretically undifferentiated community (the "peo-
ple as a whole"), may be undermined. It therefore becomes useful and
often essential for such cooptation to remain in the shadowland of in-
formal interaction.
The informal cooptation of existing nucleuses of power into the total
(formal plus informal) policy-determining structure of an organization,
symptomatic of an underlying stress, is a mechanism of adjustment to
concrete forces. On this level, interaction occurs among those who are
in a position to muster forces and make them count, which means that
the stake is a substantive reallocation of authority, rather than any
purely verbal readjustment. Formal cooptation, however, is rather more
ambiguous in relation to de facto reallocations of power. The sense of
insecurity which is interpreted by a leadership as indicating a need for
an increased sense of legitimacy in the community is a response to some-
thing generalized and diffuse. There is no hard-headed demand for a
sharing of power coming from self-conscious institutions which are in a
position to challenge the formal authority itself. The way things seem
becomes, in this context, more important than the way they are, with
the result that verbal formulas (degenerating readily into propaganda) ,
and formal organizational devices, appear to be adequate to fill the need.
The problem becomes one of manipulating public opinion, something
which is necessarily beside the point when dealing with an organized
interest group having an established and self-conscious leadership.
Formal cooptation ostensibly shares authority, but in doing so is in-
volved in a dilemma. The real point is the sharing of the public symbols
or administrative burdens of authority, and consequently public respon-
sibility, without the transfer of substantive power ; it therefore becomes
necessary to insure that the coopted elements do not get out of hand,
do not take advantage of their formal position to encroach upon the
actual arena of decision. Consequently, formal cooptation requires in-
formal control over the coopted elements lest the unity of command and
decision be imperiled. This paradox is one of the sources of persistent
tension between theory and practice in organizational behavior. The
leadership, by the very nature of its position, is committed to two con-
flicting goals : if it ignores the need for participation, the goal of coopera-
tion may be jeopardized ; if participation is allowed to go too far, the
continuity of leadership and policy may be threatened. 23
2:1 The analysis of unanticipated consequence and commitment is indispensable
to the interpretation of behavior in terms of the cooptative mechanism. The commit-
ments made in the course of action generate unanticipated consequences ; in analyzing
262 University of California Publications in Culture and Society
THE EMPIRICAL ARGUMENT RESTATED
Apart from the interest of analytical theory, the statement above ex-
plains the special focus of this inquiry and the basis for its obviously
selective approach to the TVA experience. That frame of reference has
guided the empirical analysis, of which the following is a brief recapitu-
lation.
1. The grass-roots theory became a protective ideology. In chapter
ii, an attempt was made to explain the high self -consciousness of the
TVA, as expressed in the grass-roots doctrine, on the basis of the func-
tion of that doctrine in facilitating acceptance of the Authority in its
area of operation and in fulfilling the need for some general justification
of its existence as a unique type of governmental agency. The TVA was
revolutionary both to the attitudes of local people and institutions and
to the federal governmental system. By adopting the grass-roots doc-
trine the Authority was able to stand as the champion of local institu-
tions and at the same time to devise a point of view which could be
utilized in general justification of its managerial autonomy within the
federal system. However, allegiance to this doctrine, and translation
of it into policy commitments, have created serious disaffection between
TVA and other branches of the federal government, including the De-
partment of Agriculture and the Department of the Interior. As a result,
on the basis of the TVA experience, these departments have been moved
to oppose the extension of the TVA form of organization to other areas, a
fact which is consequential for the future of the Authority itself.
2. The agricultural program was delegated 2 * to an organized admin-
istrative constituency. In the major example within TVA of grass-roots
procedure the Authority's fertilizer distribution program there was
constructed a strong constituency-relation involving the land-grant
college system on the one hand and the Agricultural Relations Depart-
ment of TVA on the other. This constituency relation may be viewed
as a case of informal cooptation, wherein strong centers of influence in
the function of these consequences we must construct a theory which will explain them
as events consistent with the needs and potentialities of the system. At the same
time, it must be understood that to formulate such defensive mechanisms as coopta-
tion is to state possible predicates. For the full understanding of organization it will
be necessary to construct a system of such relevant responses which can serve to
illuminate concrete cases.
24 Some TVA officials would question the use of "delegated" here. However, this
seems to be the most significant summary word to use, in terms of its implications.
Moreover, in his own summation of TVA policy upon the occasion of his leaving the
TVA chairmanship, David E. Lilienthal said: "The TVA has by persistent effort
delegated and thereby decentralized its functions " New York Times, November
13, 1946, p. 56.
Selznick : T VA and the Grass Roots 263
the Valley were absorbed covertly into the policy-determining structure
of the TVA. The TVA's Agricultural Relations Department assumed a
definite character, including a set of sentiments valuing the land-grant
college system as such and accepting the mission of defending that
system within the Authority. In effecting this representation, the TVA
agriculturists have been able to take advantage of the special preroga-
tives accruing to them from their formal status as an integral part of
the Authority, including the exercise of discretion within their own as-
signed jurisdiction and the exertion of pressure upon the evolution of
general policy within the Authority as a whole. The special role and
character of the TVA agricultural group limited its outlook with respect
to the participation of Negro institutions as grass-roots resources and
created a special relation to the American Farm Bureau Federation.
Yet the operation of this cooptative process probably did much to en-
hance the stability of the TVA within its area and especially to make
possible the mobilization of support in an hour of need. In this sense,
one cannot speak of the decisions which led to this situation as mistakes.
3. In a context of controversy, the TVA's commitments to its agricul-
tural constituency resulted in a factional alignment involving unantici-
pated consequences for its role on the national scene. In the exercise of
discretion in agriculture, the TVA entered a situation charged with
organizational and political conflict. The New Deal agricultural agen-
cies, such as Farm Security Administration and Soil Conservation
Service, came under attack of the powerful American Farm Bureau
Federation, which thought of them as threats to its special avenue of
access to the farm population, the extension services of the land-grant
colleges. Under the pressure of its agriculturists, the Authority did not
recognize Farm Security Administration and sought to exclude Soil
Conservation Service from operation within the Valley area. This re-
sulted in the politically paradoxical situation that the eminently New
Deal TVA failed to support agencies with which it shared a political
communion, and aligned itself with the enemies of those agencies.
4. Under the pressure of its agriculturists, the TVA gradually altered
a significant aspect of its character as a conservation agency. The TVA
agricultural group, reflecting local attitudes and interests, fought against
the policy of utilizing public ownership of land as a conservation meas-
ure and thus effectively contributed to the alteration of the initial policy
of the Authority in this respect. The issue of public ownership is taken
as character-defining in the sense that it is a focus of controversy and
division, and it was such within the TVA for an extended period. The
264 University of California Publications in Culture and Society
single-minded pursuit of its ideological and constituency interests led
the agricultural group to involve the Authority in a controversy with
the U. S. Department of the Interior over the management of TVA-
owned lands. In this matter, as in those mentioned above, the mechanics
of pressure and representation are detailed in the body of the study.
5. The grass-roots utilization of voluntary associations represents a
sharing of the burdens of and responsibility for power, rather than of
power itself. In chapter vii, the voluntary association device espe-
cially, but not exclusively, in the agricultural program is interpreted as
a case of formal cooptation, primarily for promoting organized access to
the public but also as a means of supporting the legitimacy of the TVA
program. Typically, this has meant that actual authority, and to a large
extent the organizational machinery, has been retained in the hands
of the administering agency. After nine years of operation, the county
soil associations handling TVA fertilizer were found to be still tools of
the county agent system, to which the TVA test-demonstration program
was delegated. In connection with this analysis, an operational test for
locating control over coopted citizens' groups is described, as suggested
in the question : Is approach to the association by outside elements chan-
neled through officials of the coopting agency ?
IMPLICATIONS FOR DEMOCRATIC PLANNING
In venturing this interpretation, we are pointing to some of the signifi-
cant problems which must be faced when the attempt is made to combine
democracy and planning. 23 For planning implies large-scale interven-
tion and the extended use of organizational instruments. If such concepts
as "democracy" are to be more than honorific symbols which mobilize
opinion, it is essential to make explicit the forces which will operate to
qualify and perhaps transform the democratic process.
There are three major considerations, three sources of paradox and
tension, which emerge out of this study, so far as it has implications for
democratic planning :
1. Ideologies must be seen in the context of the needs they serve ; it
is a reliable assumption that something more urgent than patterns of
belief will lie behind the strong advocacy of doctrine by an organiza-
tional leadership. The unanalyzed terms must be closely examined, for
it is also a reliable expectation that some covert adaptation in terms of
immediate necessity will have provided content for emotion-laden but
25 We speak here only of the implications of this analysis. That the grass-roots
approach has wider implications of importance for democratic planning is sufficiently
obvious.
Selznick : T VA and the Grass Roots 265
procedurally indefinite terms. It is perhaps a good rule to see whether
the ambiguities, the inherent dilemmas, of the attempt to fulfill doctrinal
demands are squarely faced ; where these are denied out of hand, the
ideological function of the doctrine is obliquely confirmed.
2. It is essential to recognize that power in a community is distributed
among those who can mobilize resources organizational, psychological,
and economic and these can effectively shape the character and role of
governmental instrumentalities. This has a dual significance. It may
result in the perversion of policy determined through representative
institutions ; and at the same time, this fact offers a tool for ensuring
the responsibility of public agencies to their client publics. Consequently,
it is naive to suppose that there is anything inherently bad in the situa-
tion wherein private organizations paralleling but independent of a
governmental administrative structure have a decisive influence on its
social policy. Again, however, the situation is inherently ambiguous.
This ambiguity must be explicitly recognized, and its mechanics under-
stood, if realistic controls are to be instituted.
3. The tendency of democratic participation to break down into ad-
ministrative involvement requires continuous attention. This must be
seen as part of the organizational problem of democracy and not as a
matter of the morals or good will of administrative agents. A realistic
examination of the factors which define formal cooptation also permits
us to make explicit precisely those points at which changes in procedure
will be effective in reinforcing meaningful democracy.
In general, therefore, we have been concerned to formulate some of
the underlying tendencies which are likely to inhibit the democratic
process. Like all conservative or pessimistic criticism, such a statement
of inherent problems seems to cast doubt upon the possibility of com-
plete democratic achievement. It does cast such a doubt. The alternative,
however, is the transformation of democracy into a Utopian notion which,
unaware of its internal dangers, is unarmed to meet them.
The TVA has been a particularly good subject for the analysis of
these problems. This is so precisely because it may be said that the
Authority has, on the whole, very effectively achieved some of its major
purposes, including the mobilization of a staff of very high quality. No
one is surprised when a weak or corrupt governmental agency does not
fulfill its doctrinal promise. When, however, a morally strong and funda-
mentally honest organization is subject to the kind of process we have
described, then the pervasive significance of that process becomes mate-
rially enhanced. In a sense, it is just because TVA is a relatively good
266 University of California Publications in Culture and Society
example of democratic administration that the evidences of weakness
in this respect are so important. It is just because the TVA stands as
something of a shining example of incorruptibility in such major mat-
ters as noncapitulation to local political interests in the hiring of per-
sonnel or to local utility interests in public power policy that the evidence
of covert cooptation in the agricultural program attains its general
significance.
For the things which are important in the analysis of democracy are
those which bind the hands of good men. We then learn that something
more than virtue is necessary in the realm of circumstance and power.
PUBLISHED SOURCES
Anderson, W. A. "Farm Youth in the 4-H Club," Department of Rural Sociology
Mimeograph Bulletin No. 14, Cornell University Agricultural Experiment Station,
Ithaca, N.Y. (May, 1944).
Association of Land-Grant Colleges and Universities. Proceedings of the 48th Annual
Convention, Washington, D.C. (1934).
Baker, Gladys. The County Agent (Chicago: University of Chicago Press, 1939).
Ball, Carleton E. "Citizens Help Plan and Operate Action Programs," Land Policy
Eeview (March-April, 1940).
. A Study of the Work of the Land-Grant Colleges in the Tennessee Valley
Area in Cooperation with the Tennessee Valley Authority, prepared under the
auspices of the Coordinating Committee of the USD A, the TVA, and Valley States
Land-Grant Colleges (No imprint, 1939).
Barnard, C. I. Dilemmas of Leadership in the Democratic Process (Princeton:
Princeton University Press, 1939).
Barnard, C. I. The Functions of the Executive (Cambridge: Harvard University
Press, 1938).
Davidson, Donald. "Political Regionalism and Administrative Regionalism," Annals
of the American Academy of Political and Social Science, Vol. 207 (January, 1941).
Gaus, J. M. "A Theory of Organization in Public Administration," The Frontiers of
Public Administration (Chicago: University of Chicago Press, 1936).
Gaus, J. M., and L. D. Wolcott. Public Administration and the U. S. Department of
Agriculture (Chicago: Public Administration Service, 1940).
Gerth, H. H., and C. Wright Mills. From Max Weber (New York: Oxford University
Press, 1946).
Gibson, D. L. "The Clientele of the Agricultural Extension Service," Michigan Agri-
cultural Experiment Station Quarterly Bulletin, Vol. 1 (May, 1944).
Gray, L. C. "Our Land Policy To-day," Land Policy Eeview, Vol. 1 (May-June, 1938) .
Gregory, Clifford V. "The American Farm Bureau Federation and the AAA," Annals
of the American Academy of Political and Social Science, Vol. 179 (May, 1935).
Gross, Neal C. "A Post Mortem on County Planning," Journal of Farm Economics,
Vol. XXV (August, 1943).
Gulick, Luther, and Lydall Urwick (eds.). Papers on the Science of Administration
(New York: Institute of Public Administration, Columbia University, 1937).
Hardin, Charles M. "The Bureau of Agricultural Economics Under Fire, A Study in
Valuation Conflicts," Journal of Farm Economics, Vol. 28 (August, 1946).
Hartford, Ellis F. Our Common Mooring (Athens: University of Georgia Press,
1941).
Hodgkin, Carlyle. "Agriculture's Got Troubles," a series in Successful Farming
(March, April, and May, 1943).
Hoffer, C. R. "Selected Social Factors Affecting Participation in Agricultural Exten-
sion Work," Michigan State College Agricultural Experiment Station Special
Bulletin 331 (June, 1944).
Iowa State College. The Role of the Land-Grant College in Governmental Agricultural
Programs (Ames: Iowa State College Bulletin, June, 1938).
Key, V. O., Jr. The Administration of Federal Grants to the States (Chicago : Public
Administration Service, 1937).
[267]
268 University of California Publications in Culture and Society
Key, V. O., Jr. "Politics and Administration," in L. D. White (ed.), The Future of
Government in the U. S. (Chicago: University of Chicago Press, 1942).
Leiserson, Avery. Administrative Regulation: A Study in the Representation of
Interests (Chicago: University of Chicago Press, 1942).
Lewis, John D. "Democratic Planning in Agriculture," American Political Science
Review, Vol. XXV (April and June, 1941).
Lilienthal, David E. "Management Responsible or Dominant?" Public Administra-
tion Review, Vol. I (Summer, 1941).
. "The TVA: An Experiment in the Grass-Eoots Administration of Federal
Functions," Address before the Southern Political Science Association, Knoxville,
Tennessee, November 10, 1939 (TVA Information Office: no date).
. TVA : Democracy on the March (New York: Pocket Books, 1945).
Lilienthal, David E., and Robert H. Marquis. "The Conduct of Business Enterprises
by the Federal Government," Harvard Law Review, Vol. 54 (February, 1941).
Lord, Russell. The Agrarian Revival (New York: American Association for Adult
Education, 1939).
Malinowski, Bronislaw. The Dynamics of Culture Change (New Haven : Yale Univer-
sity Press, 1945).
. A Scientific Theory of Culture (Chapel Hill: University of North Carolina
Press, 1944).
Merton, Robert K. "The Unanticipated Consequences of Purposive Social Action,"
American Sociological Review, Vol. I (December, 1936).
Mooney, James D., and Alan C. Reiley. The Principles of Organisation (New York:
Harpers, 1939).
New Horizons in Public Administration. A Symposium (University, Alabama: Uni-
versity of Alabama Press, 1945).
Parsons, Talcott. "The Present Position and Prospects of Systematic Theory in So-
ciology," in Georges Gurvitch and Wilbert E. Moore (eds.), Twentieth Century
Sociology (New York: Philosophical Library, 1945).
. The Structure of Social Action (New York: McGraw-Hill, 1937).
Pincus, William. "Shall We Have More TVA's?" Public Administration Review, Vol.
V (Spring, 1945).
Pritchett, C. Herman. The Tennessee Valley Authority: A Study in Public Adminis-
tration (Chapel Hill: University of North Carolina Press, 1943).
Richards, E. C. M. "The Future of TVA Forestry," Journal of Forestry, Vol. 36
(July, 1938).
Ryan, Bryce. "Democratic Telesis and County Agricultural Planning," Journal of
Farm Economics, Vol. XXII (November, 1940).
Selznick, Philip. "Foundations of the Theory of Organization," American Sociologi-
cal Review, Vol. XIII (February, 1948).
Stephens, Oren. "FSA Fights for Its Life," Harper's (April, 1943).
Tolley, Howard R. The Farmer Citizen at War (New York: Macmillan, 1943).
U. S. Bureau of Agricultural Economics. "The Land Use Planning Organization,"
County Planning Series No. 3 (May, 1940).
. Land Use Planning Under Way (Washington, D.C.: Government Printing
Office, July, 1940).
U. S. Congress. House. Committee on Appropriations. Hearings on Agriculture De-
partment Appropriation Bill for 1944, 78th Cong., 1st sess. (1943).
Selznick : T VA and the Grass Roots 269
U. S. Congress. House. Committee on Military Affairs. Muscle Shoals Hearings, 72d
Cong., 1st sess. (1932).
. Joint Committee on Eeduction of Nonessential Federal Expenditures. Hear-
ings, 77th Cong., 3d sess. (1942).
. Joint Committee to Investigate the Tennessee Valley Authority. Hearings,
75th Cong., 3d sess. (1938). Keport, 76th Cong., 1st sess. (1939).
TJ. S. Congress. Senate. Committee on Agriculture and Forestry. Hearings on S. 2555,
"To provide for the creation of conservation authorities . . . ," 75th Cong., 1st sess.
(1937).
. Senate. Committee on Agriculture and Forestry. Hearings on S. 2361, "To
amend the TVA Act of 1933, as amended, with respect to the manner of the exer-
cise of the power of condemnation by the TVA and to require the receipts of the
Authority to be covered into the Treasury," 77th Cong., 2d sess. (1942).
. Senate. Committee on Agriculture and Forestry. Hearings on S. 1251, "To
provide for the establishment of a national soil fertility policy and program . . . ,"
80th Cong., 1st sess. (1947).
. Senate. Committee on Civil Service. Hearings on H.E. 960, "An Act Extend-
ing the Classified Civil Service," 76th Cong., 3d sess. (1940).
. Senate. Committee on Commerce. Hearings on S. 555, "A Bill to Create a
Missouri Valley Authority," 79th Cong., 1st sess. (1945).
Senate. Committee on Irrigation and Eeclamation. Hearings on S. 555, "A
Bill of Create a Missouri Valley Authority," 79th Cong., 1st sess. (1945).
U. S. National Eesourees Planning Board. Public Land Acquisition in a National
Land-Use Program, Pt. I: Eural Lands (Washington, D.C.: Government Printing
Office, 1940).
Urwick, Lydall. The Elements of Administration (New York: Harpers, 1943).
Wilson, M. C. "How and to What Extent is the Extension Service Eeaching Low-
Income Farm Families?" USDA Extension Service Circular 375 (December, 1941).
Works, G. A., and Barton Morgan. The Land Grant Colleges, Staff Study No. 10,
Advisory Committee on Education (Washington, D.C. : Government Printing Of-
fice, 1939).
f
INDEX
Acceptance of TVA, 48f ., 62, 237
Access to populations
and formal cooptation, 219, 224, 233
and voluntary associations, 218, 237
Adjustment
and ideology, 48ff .
of TVA to institutional environment,
12,55
to federal system, 79
vs. symbolic import, 21
Administrative constituency, 146ff.
Agreements, administrative
and internal pressure, 96, 150
in joint agricultural program, 95 .,
124f.
reduction in meaning, 179
used as weapons, 179
Agricultural Eelations Department
and fanaticism, 210ff .
and Land Acquisition Department,
206ff.
and public ownership of land, 202, 205
as internal pressure group, 147ff. r
chaps, v, vi
character and role, lllff.
dependence on colleges, 153, 177
formal structure, 109ff .
Agriculture, U. S. Department of
and citizens' participation, 221
changes in, 170, 174
rejection of TVA as model, 78
role in memorandum of understanding,
96, 175
split in, 164
Agricultural program
extent, 99ff .
integration with extension service,
129ff.
legal basis, 89
origin, 86ff.
Alienation, 211
Association of Land-Grant Colleges and
Universities, 93f ., 96f ., 99
Baker, Gladys, 117, 120f., 157, 161
Baker, W. M., 195, 208
Ball, Carleton E., 95, 222f .
Barnard, C. I., 51, 70f ., 251
Bass, C. Niel, 105
Bell, Daniel, 185
Bennett, H. H., 171f .
Biological Survey, Bureau of, contro-
versy with TVA agriculturists, 206ff.
Bureaucracy, 9ff ., 13, 74
Burnham, James, 27
"Casual breakdown," 170
Character, organizational
and administrative discretion, 182ff .
and assumptions as to method, 209
and cooptation, 15f .
and indirect consequences, 181, 197
and indispensability of leadership, 183
and locus of decision, 210
and personnel, 32, 197
and public land policy, 188, 202f .
and values, 183
split in TVA's, 166
Chief Conservation Engineer, 105; on
land policy, 198
Chief Engineer, as ally of agriculturists,
197
Civil Service Commission, 30
Clapp, G. E., 38
Commerce Department, 108
Commitment
and cooptation, 16
and defining heroes, 69f .
and dilemmas of doctrine, 69f .
and unanticipated consequences, 73,
255ff.
in relation to adjustment, 12
professional, 75
types of, in organizational behavior,
256ff.
Community associations, 235ff.
Conf ormance, 70f .
Consequences, unanticipated, 254ff .
Conservation: early TVA goals, 190ff.;
relation of TVA to tradition of, 190,
201
Controversy: in agriculture, 157ff.; in
TVA, 54f., 92, 190, 193; historical
context of inner-organizational, 185
[271]
272
Index
Cooper, Prentice, 204
Cooper, Thomas P., 94, 173
Cooperation : as unanalyzed abstraction,
165; power implications, 166, 168
Cob'ptation: defined, 13, 259ff.; and
totalitarian-democratic continuum,
225; formal, 219ff., 226; informal,
217
Coordination: as unanalyzed abstrac-
tion, 64; utility of doctrine for, 50f.
Cooperatives : rural electrification, 238ff . ;
soil associations, 235 ; TVA interest
in, 226ff., 116
Correlating Committee: established,
95f.j limited functions, 178f.; role,
97
County agents: and voluntary associa-
tions, 231ff. (see also extension serv-
ice)
Davidson, Donald, 48
Decentralization : and regionalism, 41ff . ;
levels of, 44f.; utility of doctrine
for, 50
Decision
and analysis of commitment, 259
and administrative discretion, 66
and dilemmas of grass-roots doctrine,
69ff.
institutional context, 257
local as consequential for national, 73,
156ff.
organizational consequences of crucial,
175ff.
Delegation : and discretion, 67, 71 ;
hazards of, 25 7f.; in TVA agricul-
tural program, 262
Dewey, John, 181, 249, 254
Dickson, W. J., 251
Dilemmas: inherent in TVA doctrine,
69ff. ; in test-demonstration pro-
gram, 134ff .
Dimock, M. E., 67
Discretion, administrative
analyzed, 64ff.
and centrifugal tendencies, 155
and grass-roots doctrine, 27
and indirect planning, 6, 186f .
and organizational character, 182f .
and unanticipated consequences, 258
in land purchase, 188
Doctrine, official
administrative utility, 50
as ideology, 8
functions of TVA, 55ff.
sources of TVA, 49ff .
Douglas Dam controversy, 56, 58, 204
Draper, E. S., 192, 194
Eisenhower, Milton S., 164, 170, 174, 178
Electric power program
and grass-roots method, 57, 241f .
as justification for cob'ptation in agri-
culture, 175
implications of primacy of, 72n., 92f .,
100
rural electrification cooperatives, 238
Emerson, Howard P., 43
Extension service
character of, 117ff.
integration with TVA program, 129ff.
political potential in Tennessee Val-
ley, 184
Washington office, 96
Fanaticism, 210ff.
Farm Bureau
and access to farm population, 160ff .
and Association of Land-Grant Col-
leges, 93
and extension service, 119ff ., 160ff.
dilemma in relation to new programs,
161
role in agricultural controversy, 158ff.
role in Muscle Shoals controversy, 86ff .
TVA and, 141ff.
Federal government, relation of TVA to,
74ff.
Fertilizer : in TVA agricultural program,
86ff., 99ff.; phosphates vs. nitrates,
98f . ; test-demonstrations, 130ff .
Finer, Herman, 54
Fish and Wildlife Service (see Biological
Survey)
Forestry Eelations Department: and
Chief Conservation Engineer, 105;
efforts to reorient, 147ff.; role in
land policy, 195ff .
Gabrielson, Ira N., 206
Index
273
Gaus, J. M., 97, 160, 164, 223f .
Gray, L. C., 189
General Accounting Office, and man-
agerial autonomy, 33f .
General Manager, 115
Gesinnungsethik, 210
Grass-roots method
administrative "logic," 128
ambivalence, 211
and electrical power program, 241f .
and land-acquisition policy, 199
as sustaining loose administrative re-
lations, 124f .
Gregory, Clifford V., 161
Gross, Neal C., 225
Hardin, Charles M., 224
Harris, J. P., 96
Hartford, Ellis F., 43
Heroes, organizational, and character-
definition, 182
Herring, W. E., 238
Hodgkin, Carlyle, 159, 162
Hoskins, J. D., 61, 204
Ickes, H. L., 80
Ideological vacuum, 52, 54
Ideology: TVA doctrine as, 47ff.; and
covert representation of interests,
66
Informal structure, as guiding concept,
251
Internal relevance, as guiding concept,
252
Johnston, Oscar, 159
Key, V. O., 96f ., 117, 145f.
Laski, H. J., 67
Land Acquisition Department, and agri-
culturists, 206ff.
Land-grant colleges, 93ff.; and new
agencies, 163f. (see also extension
service)
Land-use associations, 242f .
Legitimacy of authority, 13ff ., 260
Leiserson, Avery, 66, 145
Lewis, J. D., 225f.
Lilienthal, D. E.
and states-rights interests, 60f .
as expositor of official doctrine,
22ff.
on agriculture, 116
on primacy of power program, 93
on significance of administrative meth-
ods, 75
bloc with H. A. Morgan, 92
Lord, Eussell, 117, 157, 161, 164, 224
Malaria control program, as source of
friction with Biological Survey, 207
Managerial autonomy, 29ff.j source of
TVA's, 52f .
Managerialism, 27
Mannheim, Karl, 219
Maps and Surveys Division, role in land
policy, 195
Meaning of events, 253
Merton, B. K., 69, 252, 254
Missouri Valley Authority debate, 80ff .
Moore, W. E., 251f .
Morgan, A. E., 30, 91f ., 96, 147, 172, 227
Morgan, Barton, 97, 117, 121
Morgan, H. A.
and Agricultural Eelations Depart-
ment, 114f ., 184
and decision for phosphates, 99
and Farm Bureau leaders, 143f .
and grass-roots method, 22, 43, 91ff.,
96
and land-purchase policy, 204
bloc with Lilienthal, 92
mobilizes colleges in defense of TVA,
177
Morrison, Donald, lln.
Muscle Shoals controversy, 4, 86ff .
McAmis, J. C., 95, 114, 143, 167, 173
McKellar, Kenneth, and Douglas Dam
controversy, 204
National Eesources Planning Board : and
TVA model, 78f.; on public land
policy, 189
Needs, organizational, lOff., 224f., 252,
256
Negroes, 112f.
Norris, George W., and civil service pol-
icy, 31; and farm bureau, 144
O'Neal, E. A., 143
Ortega y Gasset, Jose", 219
Parker, T. B., 197
Parsons, Talcott, 252
274
Index
Participation
and cooptation, 13f .
and democracy in administration, 220
as indirect, 74
as unanalyzed abstraction, 64
in rural electrification cooperatives,
240
reduced to involvement, 220f ., 225f .,
236f.
Pincus, William, 57
Planners, role in land policy, 192ff .
Planning : and TVA powers, 6 ; indirect,
6, 187; town planning commissions,
244ff.
Plant, James S., 170n.
Pope, James P., 204
Porter, J. F., 143
Pressure : dry method, 165, 168 ; histori-
cal context, 185; third-party tactic,
208 (see also representation of inter-
ests)
Pritchett, C. Herman, 57n.
Protective strip policy, 193ff .
Keadjustment program, 133
Regional Planning Studies Department:
criticism of agriculturists, 198 ; cur-
tailment of influence, 195; defense
of buffer belt policy, 194
Eegionalism, 5, 41ff .
Representation of interests, mechanics
of, 147ff., 206f. (see also pressure)
Reservoir Property Management Depart-
ment, 109, 168, 196
Responsibility, and formal cooptation,
225, 227, 231f.
Revenue-financing, and managerial au-
tonomy, 35f.; and organizational
character, 68
Richards, E. C. M., 147ff., 190f.
Roethlisberger, F. J., 251
Roosevelt, F. D., 5, 54
Rural Electrification Administration,
238ff.
Ryan, Bryce, 225
Soil conservation associations, cooptative
character, 231f .
Soil Conservation Service
and TVA-USDA crisis, 171ff.
and voluntary associations, 221
exclusion from Tennessee Valley,
172f.
in agricultural controversy, 163
revision of exclusion order, 178
States-rights interests, 60ff ., 72
Stephens, Oren, 159, 164
Stone, Donald C., 51n.
Structural-functional analysis, 252
Tennessee Valley Associated Coopera-
tives, 228ff.
Tennessee Valley Authority
as model, 19, 30, 78ff.
as symbol, 19, 20, 81
powers of, 6, 12, 90f ., 186f., 229
Tenancy, farm, 104, 112, 123, 138f.
Tocqueville, Alexis de, 25
Tolley, Howard R., 222
Tools of action, and commitment, 255,
258; as recalcitrant, 253
Unanalyzed abstractions, 59ff .
Uniqueness of TVA, 5, llf .
United front, 168n.
Warburton, C. W., 95
Weapons, organizational, 158, 161, 179
Weber, Max, 210
Whittington,W.M.,81
Wilson, M. L., 222
Wolcott, L. O., 97n., 160, 164, 223
Works, G. A., 97, 117, 121
V
o
<v
SELLS
PALO
A L T
Live Music Archive
Librivox Free Audio
Metropolitan Museum
Cleveland Museum of Art
Internet Arcade
Console Living Room
Books to Borrow
Open Library
TV News
Understanding 9/11